![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/bec9dbe8-c18d-4a64-a6f3-faaf7e76ef58/width=525/bec9dbe8-c18d-4a64-a6f3-faaf7e76ef58.jpeg\")
I wanted to create a model that has the quality of Flux without the limitations of NSFW content and does not require LoRa.I wanted to create a model that has the quality of Flux without the limitations of NSFW content and does not require LoRa.After a long time of work, I am now happy to be able to present my model to you.
(NSFW) Completely uncensored, allowing generation of erotic and very explicit images.
Training Configuration:
GPU: A6000
Dataset Size: 1500 images
Batch Size: 1
Optimizer: AdamW
Learning Rate: 1e-04
Total training steps: 30k
Training Duration: 17h+
Please note that all my FLUX checkpoint models are still experimental, so you might encounter some issues. If you do, please let me know in the comments section. Your feedback is valuable for future improvements!
Recommended Parameter :
Sampler : euler
Steps : 20+
CFG : 1
No VAE or CLIP is baked in. Use separate sources for those.
Minimum Parameter :
Sampler : euler
Steps : 16+
CFG : 1
LoRA's are compatible with this checkpoint if they are trained using this model Flux Dev as the base model.
LoRA's trained on this model work with other Flux Dreambooth fine tune checkpoints.
I wanted to create a model that has the quality of Flux without the limitations of NSFW content and does not require LoRa.After a long time of work, I am now happy to be able to present my model to you.
(NSFW) Completely uncensored, allowing generation of erotic and very explicit images.
Training Configuration:
GPU: A6000
Dataset Size: 1500 images
Batch Size: 1
Optimizer: AdamW
Learning Rate: 1e-04
Total training steps: 30k
Training Duration: 17h+
Please note that all my FLUX checkpoint models are still experimental, so you might encounter some issues. If you do, please let me know in the comments section. Your feedback is valuable for future improvements!
Recommended Parameter :
Sampler : euler
Steps : 20+
CFG : 1
No VAE or CLIP is baked in. Use separate sources for those.
Minimum Parameter :
Sampler : euler
Steps : 16+
CFG : 1
LoRA's are compatible with this checkpoint if they are trained using this model Flux Dev as the base model.
LoRA's trained on this model work with other Flux Dreambooth fine tune checkpoints.
No one has added the Dev model of Hi Dream to Civitai so I am uploading this here.
Original source: https://huggingface.co/Comfy-Org/HiDream-I1_ComfyUI/tree/main/split_files/diffusion_models
27/04/2025 14:15 GMT - Updated workflow to v1.1 - Better settings - Sampler now DPMPP_2M
No one has added the Dev model of Hi Dream to Civitai so I am uploading this here.
Original source: https://huggingface.co/Comfy-Org/HiDream-I1_ComfyUI/tree/main/split_files/diffusion_models
27/04/2025 14:15 GMT - Updated workflow to v1.1 - Better settings - Sampler now DPMPP_2M
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8d6d74c4-167a-463f-b789-d9d5b518cb2a/width=525/8d6d74c4-167a-463f-b789-d9d5b518cb2a.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/612961f9-f940-4f97-97b4-181c436d3590/width=525/612961f9-f940-4f97-97b4-181c436d3590.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f8e28747-aa71-4929-b62f-c1b6ca9b15f7/width=525/f8e28747-aa71-4929-b62f-c1b6ca9b15f7.jpeg\")
We didn't just upgrade it. We rebuilt it for creators, designers, and AI tinkerers who demand pixel-perfect control.
✅ Multi-Image Editing? YES.
Drag in “person + product” or “person + scene” — it blends them like magic. No more Franken-images.
✅ Single-Image? Rock-Solid Consistency.
Faces stay you — through poses, filters, and wild styles.
Products keep their identity — ideal for ads & posters.
✍️ Text? Edit everything: content, font, color, even material texture.
✅ ControlNet Built-In.
Depth. Edges. Keypoints. Plug & play precision.
This September, we are pleased to introduce Qwen-Image-Edit-2509, the monthly iteration of Qwen-Image-Edit. To experience the latest model, please visit Qwen Chat and select the \"Image Editing\" feature. Compared with Qwen-Image-Edit released in August, the main improvements of Qwen-Image-Edit-2509 include:
Multi-image Editing Support: For multi-image inputs, Qwen-Image-Edit-2509 builds upon the Qwen-Image-Edit architecture and is further trained via image concatenation to enable multi-image editing. It supports various combinations such as \"person + person,\" \"person + product,\" and \"person + scene.\" Optimal performance is currently achieved with 1 to 3 input images.
Enhanced Single-image Consistency: For single-image inputs, Qwen-Image-Edit-2509 significantly improves editing consistency, specifically in the following areas:
Improved Person Editing Consistency: Better preservation of facial identity, supporting various portrait styles and pose transformations;
Improved Product Editing Consistency: Better preservation of product identity, supporting product poster editing；
Improved Text Editing Consistency: In addition to modifying text content, it also supports editing text fonts, colors, and materials；
Native Support for ControlNet: Including depth maps, edge maps, keypoint maps, and more.
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2e1a9724-525f-49b2-9abd-3b2b5c087f4b/width=525/2e1a9724-525f-49b2-9abd-3b2b5c087f4b.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f3b70c34-d7ce-44cc-a772-a944a0669475/width=525/f3b70c34-d7ce-44cc-a772-a944a0669475.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8093de30-9f9c-4414-9db1-237eee552801/width=525/8093de30-9f9c-4414-9db1-237eee552801.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/aec184b1-00be-4393-a221-f427b619e966/width=525/aec184b1-00be-4393-a221-f427b619e966.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3061da5a-1cc6-40a2-88f3-1e22cd30bbc6/width=525/3061da5a-1cc6-40a2-88f3-1e22cd30bbc6.jpeg\")
The primary update in Qwen-Image-Edit-2509 is support for multi-image inputs.
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/23e470bd-351e-4434-bfd4-3068d8998ba4/width=525/23e470bd-351e-4434-bfd4-3068d8998ba4.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e906f90d-7220-400e-bfa8-19566fb0418e/width=525/e906f90d-7220-400e-bfa8-19566fb0418e.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/aedcc2c6-ae40-4ae6-a8fb-aabbc75a28a1/width=525/aedcc2c6-ae40-4ae6-a8fb-aabbc75a28a1.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a18fee88-8a12-44a1-8a1e-6748e0677e9d/width=525/a18fee88-8a12-44a1-8a1e-6748e0677e9d.jpeg\")
In fact, multi-image input also supports commonly used ControlNet keypoint maps—for example, changing a person’s pose:
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/bbf97199-2f51-46ec-9f31-ac11788fd3ca/width=525/bbf97199-2f51-46ec-9f31-ac11788fd3ca.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5119ca39-f1b4-4932-817c-99041c195a93/width=525/5119ca39-f1b4-4932-817c-99041c195a93.jpeg\")
Similarly, the following examples demonstrate results using three input images:
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/395c82b4-e846-4139-93e0-a85dee14cb5b/width=525/395c82b4-e846-4139-93e0-a85dee14cb5b.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/56d2b011-3bf4-42ae-a2bc-4a2e9add0e33/width=525/56d2b011-3bf4-42ae-a2bc-4a2e9add0e33.jpeg\")
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/17c61d47-849f-4187-8557-82de25ae12a9/width=525/17c61d47-849f-4187-8557-82de25ae12a9.jpeg\")
Another major update in Qwen-Image-Edit-2509 is enhanced consistency.
First, regarding person consistency, Qwen-Image-Edit-2509 shows significant improvement over Qwen-Image-Edit. Below are examples generating various portrait styles.
", - "tags": [ - "style" - ], - "allowNoCredit": false, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2138973, - "sizeKB": 19951792.125, - "name": "qwenEdit2509NewVersion_fp8E4m3fn.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-09-24T03:53:43.533Z", - "metadata": { - "format": "SafeTensor", - "size": "full", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "4E6FF885", - "AutoV2": "FA931426B3", - "SHA256": "FA931426B399E9CFCD9718607579E0271E8CC718143ACC0F186844C2A32FBD70", - "CRC32": "96F1601B", - "BLAKE3": "DAF7155D13031479A54533D91F8F7C67F2DB85863ECDCF73D90E54ACD2BF3C1A", - "AutoV3": "483010C3DA0C" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2245830" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7e0176a9-02f9-477c-b280-30c535f8eeae/original=true/102092801.jpeg", - "nsfwLevel": 1, - "width": 3780, - "height": 2268, - "hash": "UOI5JM4nI9-:_NIAjcR%%KRP-;RP%2E1RjRj", - "type": "image", - "metadata": { - "hash": "UOI5JM4nI9-:_NIAjcR%%KRP-;RP%2E1RjRj", - "size": 1463110, - "width": 3780, - "height": 2268 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2245830", - "creator": { - "username": "Cyph3r", - "image": "https://lh3.googleusercontent.com/a/AEdFTp5Qfa9qn3Bg2AxBAUviObrttQI0QGKq9_lYJ2ApEQ=s96-c" - } - }, - "tags": [ - "style" - ], - "modelDescription": "We didn't just upgrade it. We rebuilt it for creators, designers, and AI tinkerers who demand pixel-perfect control.
✅ Multi-Image Editing? YES.
Drag in “person + product” or “person + scene” — it blends them like magic. No more Franken-images.
✅ Single-Image? Rock-Solid Consistency.
Faces stay you — through poses, filters, and wild styles.
Products keep their identity — ideal for ads & posters.
✍️ Text? Edit everything: content, font, color, even material texture.
✅ ControlNet Built-In.
Depth. Edges. Keypoints. Plug & play precision.
This September, we are pleased to introduce Qwen-Image-Edit-2509, the monthly iteration of Qwen-Image-Edit. To experience the latest model, please visit Qwen Chat and select the \"Image Editing\" feature. Compared with Qwen-Image-Edit released in August, the main improvements of Qwen-Image-Edit-2509 include:
Multi-image Editing Support: For multi-image inputs, Qwen-Image-Edit-2509 builds upon the Qwen-Image-Edit architecture and is further trained via image concatenation to enable multi-image editing. It supports various combinations such as \"person + person,\" \"person + product,\" and \"person + scene.\" Optimal performance is currently achieved with 1 to 3 input images.
Enhanced Single-image Consistency: For single-image inputs, Qwen-Image-Edit-2509 significantly improves editing consistency, specifically in the following areas:
Improved Person Editing Consistency: Better preservation of facial identity, supporting various portrait styles and pose transformations;
Improved Product Editing Consistency: Better preservation of product identity, supporting product poster editing；
Improved Text Editing Consistency: In addition to modifying text content, it also supports editing text fonts, colors, and materials；
Native Support for ControlNet: Including depth maps, edge maps, keypoint maps, and more.
The primary update in Qwen-Image-Edit-2509 is support for multi-image inputs.
In fact, multi-image input also supports commonly used ControlNet keypoint maps—for example, changing a person’s pose:
Similarly, the following examples demonstrate results using three input images:
Another major update in Qwen-Image-Edit-2509 is enhanced consistency.
First, regarding person consistency, Qwen-Image-Edit-2509 shows significant improvement over Qwen-Image-Edit. Below are examples generating various portrait styles.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145824.895275, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen-image-edit/qwen_image_edit_2509_fp8_e4m3fn (20gb).png b/qwen-image-edit/qwen_image_edit_2509_fp8_e4m3fn (20gb).png deleted file mode 100644 index a94b57c574304dd7554edb1583ac4eca67ceee58..0000000000000000000000000000000000000000 --- a/qwen-image-edit/qwen_image_edit_2509_fp8_e4m3fn (20gb).png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:97a7a6297f0824c09e34fc874a904e21a3f33313d7d945eb7435c73e3ad6acf2 -size 1463110 diff --git a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.metadata.json b/qwen-image-edit/qwen_image_edit_2511_fp8mixed.metadata.json deleted file mode 100644 index 9a9249a0ddcb770338682b7193521c197f37bc6d..0000000000000000000000000000000000000000 --- a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.metadata.json +++ /dev/null @@ -1,430 +0,0 @@ -{ - "file_name": "qwen_image_edit_2511_fp8mixed", - "model_name": "Qwen-Image-Edit 2511", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen-image-edit/qwen_image_edit_2511_fp8mixed.safetensors", - "size": 20533762817, - "modified": 1769096025.123118, - "sha256": "c9fdc158e46d3b61ef75f21ae866ca2fe808bf4a53643120d1c1e87c19280a4e", - "base_model": "Qwen", - "preview_url": "C:/ComfyUI/models/diffusion_models/qwen-image-edit/qwen_image_edit_2511_fp8mixed.mp4", - "preview_nsfw_level": 1, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2532694, - "modelId": 2247803, - "name": "FP8", - "nsfwLevel": 1, - "createdAt": "2025-12-24T14:55:22.434Z", - "updatedAt": "2025-12-24T15:23:29.895Z", - "status": "Published", - "publishedAt": "2025-12-24T15:23:29.886Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Qwen", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "Comfy-Org Qwen-Image-Edit 2511 fp8mixed
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:qwen:checkpoint:civitai:2247803@2532694", - "stats": { - "downloadCount": 2373, - "thumbsUpCount": 76 - }, - "model": { - "name": "Qwen-Image-Edit 2511", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "The 2509 Workflow and LightX2v 8 Steps LoRA are compatible with this version.
LoRA: Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16 | 850MB
LoRA: Qwen-Image-Edit-2511-Lightning-4steps-V1.0-fp32 | 1.7GB
BF16 on 10GB VRAM + 64GB RAM: Workflow in the zip file. First, install SageAttention and Triton for your Python environment.( look for compatible precompiled wheels )
Online Demo | HF | GH
Significantly better overall consistency in edits, addressing drift issues.
Expanded multi-image capabilities.
Native LoRA integrations for enhanced effects like realistic lighting and viewpoint shifts.
Added support for industrial design and geometric reasoning, broadening use cases beyond basic editing.
Character Consistency: Edit a portrait to change clothing or pose while keeping the face identical.
Multi-Person Fusion: Merge two individual photos into a group shot in a shared scene.
Lighting Enhancement: Apply realistic lighting using the integrated LoRA.
Viewpoint Generation: Create new angles of a subject without retraining.
Industrial Design: Batch-generate product variations and swap materials (e.g., change metal to wood).
The model is mirrored here for convenience.
", - "tags": [ - "qwen-image", - "qwen-image-edit", - "base model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2420555, - "sizeKB": 20052502.75097656, - "name": "qwenImageEdit2511_fp8.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-12-24T15:33:30.678Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "A644EB55", - "AutoV2": "C9FDC158E4", - "SHA256": "C9FDC158E46D3B61EF75F21AE866CA2FE808BF4A53643120D1C1E87C19280A4E", - "CRC32": "2F58DDD5", - "BLAKE3": "80B367244CE366AFE49483115D32916D0C5A8BE7B94AC0F99562DCBD753FD130", - "AutoV3": "C9566B78BE61" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2532694" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/06c361c0-61ac-473b-b21a-4b4a4f055431/original=true/115035032.mp4", - "nsfwLevel": 1, - "width": 792, - "height": 1320, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 7250398, - "audio": false, - "width": 792, - "height": 1320, - "duration": 6.05 - }, - "minor": false, - "poi": false, - "meta": { - "steps": 4, - "prompt": "WAN2.2 FP16", - "sampler": "UniPC", - "cfgScale": 1 - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f0ddd567-9533-44ad-a94c-aa7511048afd/original=true/115016162.jpeg", - "nsfwLevel": 1, - "width": 792, - "height": 1320, - "hash": "UNEI|y|I[Wwd$*OXnOniK4S#WBWB,[soaeR*", - "type": "image", - "metadata": { - "hash": "UNEI|y|I[Wwd$*OXnOniK4S#WBWB,[soaeR*", - "size": 1085305, - "width": 792, - "height": 1320, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "seed": 734811405823187, - "vaes": [ - "Qwen-Image/qwen_image_vae.safetensors" - ], - "steps": 8, - "prompt": "change the background and all the yellow colors to red", - "denoise": 1, - "sampler": "Euler", - "cfgScale": 1, - "scheduler": "normal", - "additionalResources": [ - { - "name": "Qwen_Edit_2509/Qwen-Image-Edit-2509-Lightning-8steps-V1.0-bf16.safetensors", - "type": "lora", - "strength": 1 - } - ] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6cbcad68-fe0e-4fa0-b2bd-c959f340f99d/original=true/115035043.jpeg", - "nsfwLevel": 1, - "width": 760, - "height": 1368, - "hash": "USIC1?=e{1#m]-o0NGS2mlRko}xai_kCkWRj", - "type": "image", - "metadata": { - "hash": "USIC1?=e{1#m]-o0NGS2mlRko}xai_kCkWRj", - "size": 1005913, - "width": 760, - "height": 1368, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "seed": 459752819060319, - "vaes": [ - "Qwen-Image/qwen_image_vae.safetensors" - ], - "steps": 8, - "prompt": "inherit the background style of image 2 and apply it to the image one with the text in the middle", - "denoise": 1, - "sampler": "Euler", - "cfgScale": 1, - "scheduler": "normal", - "additionalResources": [ - { - "name": "Qwen-Image-Edit-2511/Qwen-Image-Edit-2511-Lightning-4steps-V1.0-fp32.safetensors", - "type": "lora", - "strength": 1 - } - ] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/935cbb47-72c8-4f66-869a-2a136f328c47/original=true/115015984.jpeg", - "nsfwLevel": 1, - "width": 768, - "height": 1280, - "hash": "UWH^:l0n}+t3--$zs*WYN1xCjFWY-nR,aLt5", - "type": "image", - "metadata": { - "hash": "UWH^:l0n}+t3--$zs*WYN1xCjFWY-nR,aLt5", - "size": 1289331, - "width": 768, - "height": 1280, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "seed": 149388583840725, - "vaes": [ - "SD3.5_vae.safetensors" - ], - "steps": 4, - "prompt": "( SD3.5 Turbo ) ### A hyper-realistic digital artwork depicting a futuristic, armored warrior with dreadlocks, covered in splattered yellow and black paint. The warrior's helmet and armor feature intricate textures, with the paint appearing wet and viscous. The background is a solid, vibrant yellow, enhancing the contrast with the black and yellow paint. ", - "denoise": 1, - "sampler": "DPM++ 2M", - "cfgScale": 1, - "scheduler": "sgm_uniform" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b5bd67ef-d944-4657-bc79-c29b7b6fb501/original=true/115016721.jpeg", - "nsfwLevel": 1, - "width": 768, - "height": 1280, - "hash": "UIBX4k%h?w.8?wIUM_oJIUM_DiRjIUs:RPM{", - "type": "image", - "metadata": { - "hash": "UIBX4k%h?w.8?wIUM_oJIUM_DiRjIUs:RPM{", - "size": 1565236, - "width": 768, - "height": 1280, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "seed": 128886119289922, - "vaes": [ - "SD3.5_vae.safetensors" - ], - "steps": 40, - "prompt": "( SD3.5 Large ) ### A close-up portrait of a ninja with spiky silver-white hair, wearing a weathered metal headband with the Uzumaki clan symbol, and a torn, frost-dusted scarf covering the lower half of his face. His left eye glows with a vivid red Sharingan, while his right eye is a calm amber. The scene is set in a snowy, desolate landscape with soft, diffused lighting. The character’s expression is intense and focused, suggesting he is preparing for battle. High detail, cinematic lighting, realistic textures, and a sense of cold, gritty atmosphere. Style: hyper-detailed anime-inspired realism, akin to official character art from Naruto: Shippuden. --ar 3:4 --v 6.0", - "denoise": 1, - "sampler": "DPM++ 2M", - "cfgScale": 4.5, - "scheduler": "sgm_uniform" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/09de576b-8209-4ae8-80bb-a06c5ba9ea18/original=true/115016926.jpeg", - "nsfwLevel": 1, - "width": 792, - "height": 1320, - "hash": "UEEo0v^l%g$~Ny57R+sm}?xYkDf+57Io$zjE", - "type": "image", - "metadata": { - "hash": "UEEo0v^l%g$~Ny57R+sm}?xYkDf+57Io$zjE", - "size": 1339533, - "width": 792, - "height": 1320, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "seed": 818749445255507, - "vaes": [ - "Qwen-Image/qwen_image_vae.safetensors" - ], - "steps": 8, - "prompt": "image 1 and image 2 are sitting together in a restaurant", - "denoise": 1, - "sampler": "Euler", - "cfgScale": 1, - "scheduler": "normal", - "additionalResources": [ - { - "name": "Qwen_Edit_2509/Qwen-Image-Edit-2509-Lightning-8steps-V1.0-bf16.safetensors", - "type": "lora", - "strength": 1 - } - ] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/057cf132-7d66-4836-8805-95990d8a4d8d/original=true/116028620.jpeg", - "nsfwLevel": 1, - "width": 1088, - "height": 1920, - "hash": "U7ExF#I3N+EyEVJR;1^Q0qNwWA$*x-EtOmV|", - "type": "image", - "metadata": { - "hash": "U7ExF#I3N+EyEVJR;1^Q0qNwWA$*x-EtOmV|", - "size": 527740, - "width": 1088, - "height": 1920, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "steps": 4, - "prompt": "夜幕下的香港，霓虹灯闪烁，热闹而温暖的街景映入眼帘，五彩缤纷的中英文招牌交相辉映，熠熠生辉。远处，一位身着赛博朋克风格露脐装的亚洲美女伫立街头。雨水洒落在人行道上，倒映着雨滴，营造出一种活泼而富有电影感的氛围。色彩鲜艳夺目，粉色、蓝色、红色和绿色交织在一起。鳞次栉比的建筑上，层层叠叠的霓虹灯招牌交相辉映，展现出浓郁的80年代香港风情。招牌上赫然写着：“榮記冰室”并附有副标题\"招牌快餐\", \"凌凌漆豬肉鋪\", \"行運茶餐廳\", \"雙刀火鷄·撒尿牛丸\", \"位元堂養陰丸\", \"堅記撒尿蝦\", \"醬爆髮型屋\"并附有副标题\"FUNKYTOWN\", \"今晚打老虎\"并附有副标题\"STEPHEN CHOW KTV\", \"奔雷手按摩\", \"龍鳳茶樓\", \"達叔雜貨鋪\", 所有的商铺灯火通明，暖黄灯光的街道。修改：将赛博朋克风格的女孩换成毛绒加菲猫面具的男人，且让这个男人站在离镜头很远的地方。", - "sampler": "Euler", - "negativePrompt": "低分辨率，人体结构错误，手部图像错误，文字，错误，手指缺失，多余手指，手指数量不足，裁剪，最差质量，低质量，正常质量，JPEG伪影，签名，水印，用户名，模糊，脚部图像错误，二维码，" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b8e483ca-a588-41af-a5a1-629f686eab7b/original=true/116029793.jpeg", - "nsfwLevel": 1, - "width": 1088, - "height": 1920, - "hash": "U9G7fBIe5K9u0sSi|;}?0.SbX4xb^#J3AHaO", - "type": "image", - "metadata": { - "hash": "U9G7fBIe5K9u0sSi|;}?0.SbX4xb^#J3AHaO", - "size": 530850, - "width": 1088, - "height": 1920, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "steps": 4, - "prompt": "夜幕下的香港，霓虹灯闪烁，热闹而温暖的街景映入眼帘，五彩缤纷的中英文招牌交相辉映，熠熠生辉。远处，一位身着赛博朋克风格露脐装的亚洲美女伫立街头。雨水洒落在人行道上，倒映着雨滴，营造出一种活泼而富有电影感的氛围。色彩鲜艳夺目，粉色、蓝色、红色和绿色交织在一起。鳞次栉比的建筑上，层层叠叠的霓虹灯招牌交相辉映，展现出浓郁的80年代香港风情。招牌上赫然写着：“榮記冰室”并附有副标题\"招牌快餐\", \"凌凌漆豬肉鋪\", \"行運茶餐廳\", \"雙刀火鷄·撒尿牛丸\", \"位元堂養陰丸\", \"堅記撒尿蝦\", \"醬爆髮型屋\"并附有副标题\"FUNKYTOWN\", \"今晚打老虎\"并附有副标题\"STEPHEN CHOW KTV\", \"奔雷手按摩\", \"龍鳳茶樓\", \"達叔雜貨鋪\", 所有的商铺灯火通明，暖黄灯光的街道。", - "sampler": "Euler", - "negativePrompt": "低分辨率，人体结构错误，手部图像错误，文字，错误，手指缺失，多余手指，手指数量不足，裁剪，最差质量，低质量，正常质量，JPEG伪影，签名，水印，用户名，模糊，脚部图像错误，二维码，" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f3218ded-de59-4f12-83e6-aa5f0b8273df/original=true/117396811.jpeg", - "nsfwLevel": 16, - "width": 2804, - "height": 4096, - "hash": "U6AS*-oe00oLF}js00Na.7j?}Ts:0gWVs8WW", - "type": "image", - "metadata": { - "hash": "U6AS*-oe00oLF}js00Na.7j?}Ts:0gWVs8WW", - "size": 16079928, - "width": 2804, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "scene from a nunsploitation b-movie filmed in 1967 " - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1a605fbe-5328-48a5-b13d-93c85a406b77/original=true/117396793.jpeg", - "nsfwLevel": 1, - "width": 2804, - "height": 4096, - "hash": "UAECLE?a5Rn4gg~pxtsm00x]={Ip_38_R%NI", - "type": "image", - "metadata": { - "hash": "UAECLE?a5Rn4gg~pxtsm00x]={Ip_38_R%NI", - "size": 18031008, - "width": 2804, - "height": 4096, - "nsfwLevelReason": "Knights Vote" - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "scene from a nunsploitation b-movie filmed in 1967 " - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2532694", - "creator": null - }, - "tags": [ - "qwen-image", - "qwen-image-edit", - "base model" - ], - "modelDescription": "The 2509 Workflow and LightX2v 8 Steps LoRA are compatible with this version.
LoRA: Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16 | 850MB
LoRA: Qwen-Image-Edit-2511-Lightning-4steps-V1.0-fp32 | 1.7GB
BF16 on 10GB VRAM + 64GB RAM: Workflow in the zip file. First, install SageAttention and Triton for your Python environment.( look for compatible precompiled wheels )
Online Demo | HF | GH
Significantly better overall consistency in edits, addressing drift issues.
Expanded multi-image capabilities.
Native LoRA integrations for enhanced effects like realistic lighting and viewpoint shifts.
Added support for industrial design and geometric reasoning, broadening use cases beyond basic editing.
Character Consistency: Edit a portrait to change clothing or pose while keeping the face identical.
Multi-Person Fusion: Merge two individual photos into a group shot in a shared scene.
Lighting Enhancement: Apply realistic lighting using the integrated LoRA.
Viewpoint Generation: Create new angles of a subject without retraining.
Industrial Design: Batch-generate product variations and swap materials (e.g., change metal to wood).
The model is mirrored here for convenience.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145826.01825, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.mp4 b/qwen-image-edit/qwen_image_edit_2511_fp8mixed.mp4 deleted file mode 100644 index 95bb193891833265b6004a090d7fa588ccbf16b1..0000000000000000000000000000000000000000 --- a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.mp4 +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bed8d6ec834bf49d4c2c46047f3e8ce8d6d0d35706a57be7f56944b97073f8a8 -size 3679769 diff --git a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.png b/qwen-image-edit/qwen_image_edit_2511_fp8mixed.png deleted file mode 100644 index ed8aa92721b202769e4bb60ac60ed7e4c026473b..0000000000000000000000000000000000000000 --- a/qwen-image-edit/qwen_image_edit_2511_fp8mixed.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9d03351c7b1efa8ac69769c230867be4038feac5edfb21e0460e86691b4c2993 -size 7250398 diff --git a/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.metadata.json b/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.metadata.json deleted file mode 100644 index 46ec8c4225fe33774c4f6ffe982e0c2b1c8c2be7..0000000000000000000000000000000000000000 --- a/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.metadata.json +++ /dev/null @@ -1,351 +0,0 @@ -{ - "file_name": "qwen_image_edit_fp8_e4m3fn", - "model_name": "Qwen-Image-Edit FP8 & GGUF", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.safetensors", - "size": 20430635136, - "modified": 1769096048.645074, - "sha256": "393c6743d1de2e9031b5197027b36116f2096958ccc0223526d34e1860266021", - "base_model": "Qwen", - "preview_url": "C:/ComfyUI/models/diffusion_models/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.png", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2129225, - "modelId": 1881174, - "name": "FP8-e4m3fn", - "nsfwLevel": 3, - "createdAt": "2025-08-19T14:32:09.545Z", - "updatedAt": "2025-08-24T18:24:00.114Z", - "status": "Published", - "publishedAt": "2025-08-19T15:08:30.036Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Qwen", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": null, - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:qwen:checkpoint:civitai:1881174@2129225", - "stats": { - "downloadCount": 1574, - "thumbsUpCount": 62 - }, - "model": { - "name": "Qwen-Image-Edit FP8 & GGUF", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Error: Can't find mmproj file for 'Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf' (matching:'qwen2.5-vl-7b-instruct')! Qwen-Image-Edit will be broken!
Rename the mmproj-F16:
![](\"https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5359bf23-cc7b-488e-a784-8fdd884ce37e/width=525/5359bf23-cc7b-488e-a784-8fdd884ce37e.jpeg\")
Demo: https://chat.qwen.ai/
Text encooder FP8: https://civitai.com/models/1864281?modelVersionId=2110105
Workflow GGUF: https://civitai.com/models/1881851/qwen-iamge-edit-gguf-lightning-4steps-comfyui
This FP8-optimized (e4m3fn) conversion of Qwen-Image-Edit, now fully compatible with the latest Qwen-VL image editing features! Leveraging Alibaba's advanced multimodal AI, this model enables seamless text-guided image manipulation, perfect for precise edits, prompt-based adjustments, and creative refinements. The model is mirrored here for convenience.
", - "tags": [ - "qwen", - "qwen edit", - "base model", - "qwen image" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2023275, - "sizeKB": 19951792.125, - "name": "qwenImageEditFP8GGUF_fp8E4m3fn.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-08-19T15:12:39.106Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "041464E2", - "AutoV2": "393C6743D1", - "SHA256": "393C6743D1DE2E9031B5197027B36116F2096958CCC0223526D34E1860266021", - "CRC32": "AE0D0120", - "BLAKE3": "243C5F1F341CFD2BB59458AC3BE3029394DC5885CF691A96F7D3390A75AA4D5A", - "AutoV3": "DCAB6CAE476E" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2129225" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0dc799c2-8ca8-4528-a898-7d004a62fedc/original=true/95236859.jpeg", - "nsfwLevel": 1, - "width": 768, - "height": 1152, - "hash": "UVF6-ARPWAof_4WBWBj[?HayWVj]adjsj[f7", - "type": "image", - "metadata": { - "hash": "UVF6-ARPWAof_4WBWBj[?HayWVj]adjsj[f7", - "size": 90105, - "width": 768, - "height": 1152 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "change the text \"AuRa\" to \"Qwen\" and maintain the design and typography." - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d672205b-c5f6-4e7a-80b6-4a8a6a2c156a/original=true/95236888.jpeg", - "nsfwLevel": 1, - "width": 832, - "height": 1248, - "hash": "UTFZmgM{V?of_4WCWBay?Gj[WXkCjEf5oLoe", - "type": "image", - "metadata": { - "hash": "UTFZmgM{V?of_4WCWBay?Gj[WXkCjEf5oLoe", - "size": 847304, - "width": 832, - "height": 1248 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "change the text \"AuRa\" to \"Qwen\" and maintain the design and typography." - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e2d5919f-6762-4545-8671-b34454a0a7ad/original=true/95438483.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UEH24ex[?^=|00ob00IqyDD%={Si.7X5VrVs", - "type": "image", - "metadata": { - "hash": "UEH24ex[?^=|00ob00IqyDD%={Si.7X5VrVs", - "size": 6751552, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/899ce5f2-9845-43c0-82b3-d02de320506a/original=true/95438482.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UBHUwk00_N%200In00-;?v?HxZ-;b^4oxDNI", - "type": "image", - "metadata": { - "hash": "UBHUwk00_N%200In00-;?v?HxZ-;b^4oxDNI", - "size": 6858080, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/95d5aec5-4928-4b28-b59a-dcd0efc60422/original=true/95438484.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UFD+l7019u^*~VD%WAtm-;Mx%1T0x]M{slbI", - "type": "image", - "metadata": { - "hash": "UFD+l7019u^*~VD%WAtm-;Mx%1T0x]M{slbI", - "size": 6874140, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a56026b9-839c-49fe-be93-5492e1084695/original=true/95438513.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UDF=gh?]OWH=M_bb9ZIU%z%M-oIp-;x]nNIU", - "type": "image", - "metadata": { - "hash": "UDF=gh?]OWH=M_bb9ZIU%z%M-oIp-;x]nNIU", - "size": 6984784, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d31de65d-4e65-4478-ac20-f03c62cc987c/original=true/95438522.jpeg", - "nsfwLevel": 4, - "width": 1664, - "height": 2304, - "hash": "UWG*8CNgD]%0-HxbxJWC0gWB${WB9qRjR}jZ", - "type": "image", - "metadata": { - "hash": "UWG*8CNgD]%0-HxbxJWC0gWB${WB9qRjR}jZ", - "size": 6124137, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/eb12cde1-465d-428c-a8af-3e2b46c4814d/original=true/95438523.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UVGHY@XUDuxb^BkX%3WGN2WFW-WC0xRkt0ou", - "type": "image", - "metadata": { - "hash": "UVGHY@XUDuxb^BkX%3WGN2WFW-WC0xRkt0ou", - "size": 6015810, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6a750473-c907-456f-a72d-88d951b3f0d6/original=true/95438524.jpeg", - "nsfwLevel": 1, - "width": 1664, - "height": 2304, - "hash": "UaGHV:o}EBRo^UkDn:WY9cWC$~oe55V@s%k9", - "type": "image", - "metadata": { - "hash": "UaGHV:o}EBRo^UkDn:WY9cWC$~oe55V@s%k9", - "size": 6198404, - "width": 1664, - "height": 2304 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ac4cecc4-d0bd-425d-912e-1e480c174d69/original=true/95473474.jpeg", - "nsfwLevel": 2, - "width": 832, - "height": 1248, - "hash": "UDBDmNE1T#Imx^Q+M{%1oaj]Mcs;.TEkS3xu", - "type": "image", - "metadata": { - "hash": "UDBDmNE1T#Imx^Q+M{%1oaj]Mcs;.TEkS3xu", - "size": 1142724, - "width": 832, - "height": 1248, - "skipScannedAtReassignment": true - }, - "minor": false, - "poi": false, - "meta": { - "seed": 728943073295456, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 4, - "prompt": "change the helmet color to highly reflective black", - "denoise": 1, - "sampler": "Euler", - "cfgScale": 1, - "scheduler": "simple", - "negativePrompt": " ignore", - "additionalResources": [ - { - "name": "Qwen/Qwen-Image-Lightning-4steps-V1.0.safetensors", - "type": "lora", - "strength": 0.9000000000000001 - } - ] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2129225", - "creator": null - }, - "tags": [ - "qwen", - "qwen edit", - "base model", - "qwen image" - ], - "modelDescription": "Error: Can't find mmproj file for 'Qwen2.5-VL-7B-Instruct-Q3_K_M.gguf' (matching:'qwen2.5-vl-7b-instruct')! Qwen-Image-Edit will be broken!
Rename the mmproj-F16:
Demo: https://chat.qwen.ai/
Text encooder FP8: https://civitai.com/models/1864281?modelVersionId=2110105
Workflow GGUF: https://civitai.com/models/1881851/qwen-iamge-edit-gguf-lightning-4steps-comfyui
This FP8-optimized (e4m3fn) conversion of Qwen-Image-Edit, now fully compatible with the latest Qwen-VL image editing features! Leveraging Alibaba's advanced multimodal AI, this model enables seamless text-guided image manipulation, perfect for precise edits, prompt-based adjustments, and creative refinements. The model is mirrored here for convenience.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145826.750315, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.png b/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.png deleted file mode 100644 index 27f92147027009032fa131814161bdf03d79bd48..0000000000000000000000000000000000000000 Binary files a/qwen-image-edit/qwen_image_edit_fp8_e4m3fn.png and /dev/null differ diff --git a/qwen/Qwen-Rapid-AIO-NSFW-v22.metadata.json b/qwen/Qwen-Rapid-AIO-NSFW-v22.metadata.json deleted file mode 100644 index 6dd6253cb3354f2d90f08b58d2495dac3088eda9..0000000000000000000000000000000000000000 --- a/qwen/Qwen-Rapid-AIO-NSFW-v22.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Qwen-Rapid-AIO-NSFW-v22", - "model_name": "Qwen-Rapid-AIO-NSFW-v22", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen/Qwen-Rapid-AIO-NSFW-v22.safetensors", - "size": 28431839775, - "modified": 1769095854.410331, - "sha256": "4e8d7689ceaca6e60305c07c3cfb697a70b312da2d3827598c7e932930da3362", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145827.004738, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen/Qwen-Rapid-AIO-NSFW-v5.3.metadata.json b/qwen/Qwen-Rapid-AIO-NSFW-v5.3.metadata.json deleted file mode 100644 index f74db474c02c42423ab968fe899a9337139af2fc..0000000000000000000000000000000000000000 --- a/qwen/Qwen-Rapid-AIO-NSFW-v5.3.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Qwen-Rapid-AIO-NSFW-v5.3", - "model_name": "Qwen-Rapid-AIO-NSFW-v5.3", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen/Qwen-Rapid-AIO-NSFW-v5.3.safetensors", - "size": 28978664066, - "modified": 1769095886.215087, - "sha256": "9ee5f6b05158e9a9d203a7fe6b68c267d90d77e19b132032c1caded1f4508c12", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145827.250633, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen/qwen-image-lightningv1.1-8steps (svdq-int4_r32-).metadata.json b/qwen/qwen-image-lightningv1.1-8steps (svdq-int4_r32-).metadata.json deleted file mode 100644 index 7fa0d73b28d0d2c376ca77439963e2b9d3f33e92..0000000000000000000000000000000000000000 --- a/qwen/qwen-image-lightningv1.1-8steps (svdq-int4_r32-).metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "qwen-image-lightningv1.1-8steps (svdq-int4_r32-)", - "model_name": "qwen-image-lightningv1.1-8steps (svdq-int4_r32-)", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen/qwen-image-lightningv1.1-8steps (svdq-int4_r32-).safetensors", - "size": 11521979944, - "modified": 1769095801.103135, - "sha256": "7d17b813a2ac7217d9b55cdc3b893c545add12f8f42d8e32c1daa546b1cc3e9e", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145827.487125, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen/qwen_image_fp8_e4m3fn.jpeg b/qwen/qwen_image_fp8_e4m3fn.jpeg deleted file mode 100644 index 385be6ae614bbb46103bde3db297eefa721af5a1..0000000000000000000000000000000000000000 Binary files a/qwen/qwen_image_fp8_e4m3fn.jpeg and /dev/null differ diff --git a/qwen/qwen_image_fp8_e4m3fn.metadata.json b/qwen/qwen_image_fp8_e4m3fn.metadata.json deleted file mode 100644 index 23fc5d9a9f4fe747aaa6e0e671f23a998a0dc98b..0000000000000000000000000000000000000000 --- a/qwen/qwen_image_fp8_e4m3fn.metadata.json +++ /dev/null @@ -1,421 +0,0 @@ -{ - "file_name": "qwen_image_fp8_e4m3fn", - "model_name": "Qwen_Image_fp8_e4m3fn", - "file_path": "C:/ComfyUI/models/diffusion_models/qwen/qwen_image_fp8_e4m3fn.safetensors", - "size": 20430635136, - "modified": 1769095908.69138, - "sha256": "98763a127701eb6fb59096f7742cb3aa7d64ed510b9f4e882d8351f8176e3ce3", - "base_model": "Qwen", - "preview_url": "C:/ComfyUI/models/diffusion_models/qwen/qwen_image_fp8_e4m3fn.jpeg", - "preview_nsfw_level": 1, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2086298, - "modelId": 1843568, - "name": "v1.0", - "nsfwLevel": 3, - "createdAt": "2025-08-06T06:08:00.413Z", - "updatedAt": "2025-08-12T08:18:04.883Z", - "status": "Published", - "publishedAt": "2025-08-06T07:42:35.243Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Qwen", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": null, - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:qwen:checkpoint:civitai:1843568@2086298", - "stats": { - "downloadCount": 4535, - "thumbsUpCount": 328 - }, - "model": { - "name": "Qwen_Image_fp8_e4m3fn", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Qwen-Image is a powerful multimodal model developed by Alibaba Cloud, designed to understand and generate both text and images. It supports high-resolution image generation and complex visual reasoning, making it suitable for a wide range of AI applications. As part of the Qwen family, Qwen-Image integrates seamlessly with other models to enhance creative and analytical tasks across modalities.
", - "tags": [ - "qwen", - "base model", - "qwen_image_fp8_e4m3fn", - "fp8", - "model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "RentCivit", - "Rent" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 1982308, - "sizeKB": 19951792.125, - "name": "qwenImageFp8E4m3fn_v10.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-08-06T07:49:08.308Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "755782B1", - "AutoV2": "98763A1277", - "SHA256": "98763A127701EB6FB59096F7742CB3AA7D64ED510B9F4E882D8351F8176E3CE3", - "CRC32": "D40C9B80", - "BLAKE3": "E792E6D19BED5838E71EE3947C0F10AF0D188E487FD79D93C3B312E0433AB308", - "AutoV3": "57DB814A57D4" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2086298" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/848d09ec-3c0c-44c1-b258-f5fb1eeffd89/original=true/92612788.jpeg", - "nsfwLevel": 1, - "width": 1328, - "height": 1328, - "hash": "U7CqPunU0{aL$lFaEeah9bk6EyS~}Xw4oIs+", - "type": "image", - "metadata": { - "hash": "U7CqPunU0{aL$lFaEeah9bk6EyS~}Xw4oIs+", - "size": 2701033, - "width": 1328, - "height": 1328 - }, - "minor": false, - "poi": false, - "meta": { - "seed": 360061445559465, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 20, - "width": 1328, - "height": 1328, - "models": [], - "prompt": "\"A vibrant, warm neon-lit street scene in Hong Kong at the afternoon, with a mix of colorful Chinese and English signs glowing brightly. The atmosphere is lively, cinematic, and rain-washed with reflections on the pavement. The colors are vivid, full of pink, blue, red, and green hues. Crowded buildings with overlapping neon signs. 1980s Hong Kong style. Signs include:\n\"龍鳳冰室\" \"金華燒臘\" \"HAPPY HAIR\" \"鴻運茶餐廳\" \"EASY BAR\" \"永發魚蛋粉\" \"添記粥麵\" \"SUNSHINE MOTEL\" \"美都餐室\" \"富記糖水\" \"太平館\" \"雅芳髮型屋\" \"STAR KTV\" \"銀河娛樂城\" \"百樂門舞廳\" \"BUBBLE CAFE\" \"萬豪麻雀館\" \"CITY LIGHTS BAR\" \"瑞祥香燭莊\" \"文記文具\" \"GOLDEN JADE HOTEL\" \"LOVELY BEAUTY\" \"合興百貨\" \"興旺電器\" And the background is warm yellow street and with all stores' lights on.", - "denoise": 1, - "sampler": "Euler", - "cfgScale": 2.5, - "modelIds": [], - "scheduler": "simple", - "upscalers": [], - "versionIds": [], - "controlNets": [], - "additionalResources": [] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1378461b-7f9f-43e7-8026-ac96abc969ae/original=true/92612895.jpeg", - "nsfwLevel": 2, - "width": 768, - "height": 1152, - "hash": "U8D]A1Vz00IS?dELMy$*0dKHXkIUNXSHsjE2", - "type": "image", - "metadata": { - "hash": "U8D]A1Vz00IS?dELMy$*0dKHXkIUNXSHsjE2", - "size": 1362080, - "width": 768, - "height": 1152 - }, - "minor": false, - "poi": false, - "meta": { - "seed": 856203716880021, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 14, - "width": 768, - "height": 1152, - "models": [], - "prompt": "an ultra-high quality photo. photorealistic. Create a vibrant and surreal digital artwork featuring a sexy 22 years old japanese woman with long, flowing hair that transitions from white at the roots to vivid shades of blue, pink, and purple. deep cleavage. huge breasts. She is adorned in an elaborate, avant-garde outfit made up of layered, textured fabrics and materials, including torn papers, sketches, and abstract patterns. The clothing appears to be a chaotic yet harmonious collage of colors, shapes, and designs, giving it a dynamic and artistic feel. The background is filled with swirling, circular patterns and fragmented elements, enhancing the overall sense of movement and complexity. The overall aesthetic should blend elements of fashion, art, and fantasy, creating a visually stunning and thought-provoking piece. she is smirking.\nrandom hand posture. random posture.", - "denoise": 1, - "sampler": "res_multistep", - "cfgScale": 1, - "modelIds": [], - "scheduler": "simple", - "upscalers": [], - "versionIds": [], - "controlNets": [], - "additionalResources": [] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/bb9f3342-cc61-4452-a85b-3f6a6b867668/original=true/92612899.jpeg", - "nsfwLevel": 2, - "width": 896, - "height": 1328, - "hash": "U6BpFNEL009a.TEM$#ni0fD%%L-o%gVYjXD*", - "type": "image", - "metadata": { - "hash": "U6BpFNEL009a.TEM$#ni0fD%%L-o%gVYjXD*", - "size": 1765867, - "width": 896, - "height": 1328 - }, - "minor": false, - "poi": false, - "meta": { - "seed": 519808976118045, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 15, - "width": 896, - "height": 1328, - "models": [], - "prompt": "an ultra-high quality photo. photorealistic. Create a vibrant and surreal digital artwork featuring a sexy 22 years old japanese woman with intircate floral and ornate accessories. deep cleavage. Her attire is a bold, avant-garde ensemble that combines elements of fashion, art, and fantasy. The outfit is heavily detailed with layered textures, patterns, and vibrant colors, including metallic accents, floral designs, and abstract shapes. Her body is decorated with elaborate tattoos or painted motifs that blend seamlessly with her clothing, creating a cohesive and eye-catching look. The background is filled with swirling, organic patterns and fragmented elements, adding depth and movement to the composition. The overall aesthetic should be a fusion of elegance, creativity, and otherworldly beauty, capturing a sense of mystery and intrigue.\nrandom hand posture. random posture.", - "denoise": 1, - "sampler": "res_multistep", - "cfgScale": 1, - "modelIds": [], - "scheduler": "simple", - "upscalers": [], - "versionIds": [], - "controlNets": [], - "additionalResources": [] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f417e369-0ec4-412a-8a16-000655e253ea/original=true/92621094.jpeg", - "nsfwLevel": 2, - "width": 1136, - "height": 1472, - "hash": "UFKSJ4.8Pq$P~Xs.E1NF56s9MvEL#VxHX5wK", - "type": "image", - "metadata": { - "hash": "UFKSJ4.8Pq$P~Xs.E1NF56s9MvEL#VxHX5wK", - "size": 1737186, - "width": 1136, - "height": 1472 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "a 22 years old japanese_woman cosplaying as Asuka Langley Soryu from Neon Genesis Evangelion. She has long, bright blonde hair styled into twin ponytails. The background is a cozy, anime-themed room with pink walls adorned with framed posters of popular characters from the series. Shelves display an array of colorful figurines, including Evangelion units and other recognizable characters, adding depth and context to the scene. The lighting is soft yet vivid, highlighting the glossy texture of the costume and creating a lively, immersive atmosphere." - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/01c2640a-4709-448c-8d8a-023789a1065f/original=true/92622041.jpeg", - "nsfwLevel": 2, - "width": 1136, - "height": 1472, - "hash": "UQLM0Xtm^+IT~VE1InNZIpxXITtRrsSKRPxa", - "type": "image", - "metadata": { - "hash": "UQLM0Xtm^+IT~VE1InNZIpxXITtRrsSKRPxa", - "size": 1705076, - "width": 1136, - "height": 1472 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "a 22 years old japanese_woman cosplaying as Asuka Langley Soryu from Neon Genesis Evangelion. She has long, bright blonde hair styled into twin ponytails. The background is a cozy, anime-themed room with pink walls adorned with framed posters of popular characters from the series. Shelves display an array of colorful figurines, including Evangelion units and other recognizable characters, adding depth and context to the scene. The lighting is soft yet vivid, highlighting the glossy texture of the costume and creating a lively, immersive atmosphere." - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/77327d1c-5b4b-4274-9d22-c8ce92531c3e/original=true/92622184.jpeg", - "nsfwLevel": 1, - "width": 1136, - "height": 1472, - "hash": "UEEel.Mg;2k??dax-WShMz#-=$Oq57xWE1$*", - "type": "image", - "metadata": { - "hash": "UEEel.Mg;2k??dax-WShMz#-=$Oq57xWE1$*", - "size": 1881221, - "width": 1136, - "height": 1472 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2ce6ff7e-eee9-4ad8-9ddb-67d09ed2d560/original=true/92622352.jpeg", - "nsfwLevel": 2, - "width": 896, - "height": 1328, - "hash": "U8EVmYD%ysbq00Z~00obxuxuTHksO?EKMy-q", - "type": "image", - "metadata": { - "hash": "U8EVmYD%ysbq00Z~00obxuxuTHksO?EKMy-q", - "size": 1740733, - "width": 896, - "height": 1328 - }, - "minor": false, - "poi": false, - "meta": { - "seed": 759680011229795, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 15, - "width": 896, - "height": 1328, - "models": [], - "prompt": "an ultra-high quality photo. photorealistic. Create a vibrant and surreal digital artwork featuring a sexy 22 years old japanese woman with vibrant, multicolored hair that transitions from bright pink to deep blue. gigantic_breasts. beautiful legs. Her hair is adorned with an elaborate and whimsical headpiece made of feathers, flowers, and intricate details, giving her a fantastical and ethereal appearance. She is wearing a bold, avant-garde outfit that combines elements of gothic fashion with futuristic design. The clothing feature intricate patterns, metallic accents, and detailed textures, including swirling motifs and abstract designs. Her body is decorated with striking tattoos or painted patterns that blend seamlessly with her attire, adding depth and complexity to the overall look. The background is filled with dynamic, flowing lines and organic shapes, enhancing the surreal and artistic atmosphere. The overall aesthetic should be a fusion of elegance, creativity, and otherworldly beauty, capturing a sense of confidence and individuality.\nrandom hand posture. random posture.", - "denoise": 1, - "sampler": "res_multistep", - "cfgScale": 1, - "modelIds": [], - "scheduler": "simple", - "upscalers": [], - "versionIds": [], - "controlNets": [], - "negativePrompt": "watermark, blurry, low quality, bad quality.", - "additionalResources": [] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d8476aa1-76bd-4086-8d35-0510e8f999f7/original=true/92987630.jpeg", - "nsfwLevel": 2, - "width": 928, - "height": 1664, - "hash": "U6GIDp5S00$jHqNGwbE2%zIA~q9]00-=NH9E", - "type": "image", - "metadata": { - "hash": "U6GIDp5S00$jHqNGwbE2%zIA~q9]00-=NH9E", - "size": 87582, - "width": 928, - "height": 1664 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "an ultra-high quality photo. photorealistic. a 22 years old japanese_woman. beautiful legs. deep_cleavage. she has huge boobs.she has huge chest. she is looking at the camera. no depth of field. Side-Swept Curls hairstyle.\nrandom hand posture. random posture. no simple background. cute outfit. necklace. very short pleated_skirt. full_body. high_heels." - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/24194c7c-96f2-4c24-98c0-3159e19ddf30/original=true/94351971.jpeg", - "nsfwLevel": 1, - "width": 768, - "height": 1152, - "hash": "UGDvZe*0Ip$K9ukDS#V@ibivWVo#4nI9%1o~", - "type": "image", - "metadata": { - "hash": "UGDvZe*0Ip$K9ukDS#V@ibivWVo#4nI9%1o~", - "size": 1073752, - "width": 768, - "height": 1152 - }, - "minor": false, - "poi": false, - "meta": { - "seed": 18884891654257, - "vaes": [ - "qwen_image_vae.safetensors" - ], - "steps": 4, - "width": 768, - "height": 1152, - "prompt": "猫,computer,", - "denoise": 0.7, - "sampler": "Euler", - "cfgScale": 1, - "scheduler": "simple", - "additionalResources": [ - { - "name": "Qwen-Image-Lightning-4steps-V1.0.safetensors", - "type": "lora", - "strength": 1 - }, - { - "name": "adorablegirls.safetensors", - "type": "lora", - "strength": 0.5 - }, - { - "name": "lenovo.safetensors", - "type": "lora", - "strength": 0.8 - } - ] - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2086298", - "creator": { - "username": "aigirlfriend555", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c5551200-07be-46da-b4f3-77b93e02d2b3/width=96/aigirlfriend555.jpeg" - } - }, - "tags": [ - "qwen", - "base model", - "qwen_image_fp8_e4m3fn", - "fp8", - "model" - ], - "modelDescription": "Qwen-Image is a powerful multimodal model developed by Alibaba Cloud, designed to understand and generate both text and images. It supports high-resolution image generation and complex visual reasoning, making it suitable for a wide range of AI applications. As part of the Qwen family, Qwen-Image integrates seamlessly with other models to enhance creative and analytical tasks across modalities.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145827.771567, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/qwen/qwen_image_fp8_e4m3fn.png b/qwen/qwen_image_fp8_e4m3fn.png deleted file mode 100644 index 85d6826f4b310a62de50bb91147343cb039ea969..0000000000000000000000000000000000000000 --- a/qwen/qwen_image_fp8_e4m3fn.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3dac198e1650663d7e47b4e819325d3cb3e28258085858f169fa39ce27da3b7e -size 2701033 diff --git a/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.gguf b/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.gguf deleted file mode 100644 index f12777055cd45ef1fa692a0b1d6120be8c1510c5..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.gguf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8426c0c02a2878f4f20c28a620a1516787045a7a56301f5b6a1d91198c15a8c1 -size 5300957696 diff --git a/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.metadata.json b/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.metadata.json deleted file mode 100644 index 1d4c680359ea26797cb4ec9bb34e006e56822b12..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-I2V-A14B-HighNoise-Q2_K", - "model_name": "Wan2.2-I2V-A14B-HighNoise-Q2_K", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-I2V-A14B-HighNoise-Q2_K.gguf", - "size": 5300957696, - "modified": 1769096086.932612, - "sha256": "8426c0c02a2878f4f20c28a620a1516787045a7a56301f5b6a1d91198c15a8c1", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145838.509518, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/Wan2.2-I2V-A14B-HighNoise-Q4_K_S.metadata.json b/wan/Wan2.2-I2V-A14B-HighNoise-Q4_K_S.metadata.json deleted file mode 100644 index ff49a6250f01ed419e1748c576953141f47f7d3c..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-HighNoise-Q4_K_S.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-I2V-A14B-HighNoise-Q4_K_S", - "model_name": "Wan2.2-I2V-A14B-HighNoise-Q4_K_S", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-I2V-A14B-HighNoise-Q4_K_S.gguf", - "size": 8748151296, - "modified": 1769096096.903677, - "sha256": "8fc5f726ccf1214e43c6a599bdef1e63daf7fab06e71fb74081416c4b9ce200b", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145838.784667, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.gguf b/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.gguf deleted file mode 100644 index 045038f13f74200615b4262dd7077d376f482966..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.gguf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:384b33bfebf1e3277abc2373b9d55d9672e49f3cfa92887161a6c72dd479ec5d -size 5300957696 diff --git a/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.metadata.json b/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.metadata.json deleted file mode 100644 index c33ab1e27c5bc1b18df00c9313daa5542e62f774..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-I2V-A14B-LowNoise-Q2_K", - "model_name": "Wan2.2-I2V-A14B-LowNoise-Q2_K", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-I2V-A14B-LowNoise-Q2_K.gguf", - "size": 5300957696, - "modified": 1769096102.740715, - "sha256": "384b33bfebf1e3277abc2373b9d55d9672e49f3cfa92887161a6c72dd479ec5d", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145839.091017, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.gguf b/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.gguf deleted file mode 100644 index b597a37ade7a71a81466f388acc797f65d3cbeca..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.gguf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:593a54da69990d208dcdce18d979e41faa9ca960ae37e3e535a6ce344fdb46f9 -size 8558096896 diff --git a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.metadata.json b/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.metadata.json deleted file mode 100644 index 38046ad9dfc3f93f8923cf9567e9511b8876c2ea..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-I2V-A14B-LowNoise-Q4_0", - "model_name": "Wan2.2-I2V-A14B-LowNoise-Q4_0", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-I2V-A14B-LowNoise-Q4_0.gguf", - "size": 8558096896, - "modified": 1769096112.421869, - "sha256": "593a54da69990d208dcdce18d979e41faa9ca960ae37e3e535a6ce344fdb46f9", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145839.306484, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_K_S.metadata.json b/wan/Wan2.2-I2V-A14B-LowNoise-Q4_K_S.metadata.json deleted file mode 100644 index 05f79a52571983f987bea37274073ecf6653ae44..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-I2V-A14B-LowNoise-Q4_K_S.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-I2V-A14B-LowNoise-Q4_K_S", - "model_name": "Wan2.2-I2V-A14B-LowNoise-Q4_K_S", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-I2V-A14B-LowNoise-Q4_K_S.gguf", - "size": 8748151296, - "modified": 1769096121.996923, - "sha256": "54e2313ef7fbb169e72ab1d7cecdc90a04db01097c86f9f39097b8caacea50fb", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145839.614978, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/Wan2.2-TI2V-5B-Q4_K_M.metadata.json b/wan/Wan2.2-TI2V-5B-Q4_K_M.metadata.json deleted file mode 100644 index 396cf24288e822746bae5613aa080d13c741eedf..0000000000000000000000000000000000000000 --- a/wan/Wan2.2-TI2V-5B-Q4_K_M.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "Wan2.2-TI2V-5B-Q4_K_M", - "model_name": "Wan2.2-TI2V-5B-Q4_K_M", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/Wan2.2-TI2V-5B-Q4_K_M.gguf", - "size": 3433116000, - "modified": 1769096125.734409, - "sha256": "95b19697b7f98e65b0a543640e9ca7b4dfec32e2a6e3731e8e10708be52655e2", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145839.920513, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.1-i2v-14b-480p-Q3_K_S.metadata.json b/wan/wan2.1-i2v-14b-480p-Q3_K_S.metadata.json deleted file mode 100644 index b870f4a0542caf39179e3c1405134ccfc8a60aab..0000000000000000000000000000000000000000 --- a/wan/wan2.1-i2v-14b-480p-Q3_K_S.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "wan2.1-i2v-14b-480p-Q3_K_S", - "model_name": "wan2.1-i2v-14b-480p-Q3_K_S", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.1-i2v-14b-480p-Q3_K_S.gguf", - "size": 7925939584, - "modified": 1769096063.369426, - "sha256": "ed6dc77278b1d95afa36fa5ccead05335779ea18c5ada5eb74e61dfffb56a003", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145840.166949, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.1-i2v-14b-720p-Q3_K_S.metadata.json b/wan/wan2.1-i2v-14b-720p-Q3_K_S.metadata.json deleted file mode 100644 index c33522628f0de9d6ffd5139b343db7ebf2687ef7..0000000000000000000000000000000000000000 --- a/wan/wan2.1-i2v-14b-720p-Q3_K_S.metadata.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "file_name": "wan2.1-i2v-14b-720p-Q3_K_S", - "model_name": "wan2.1-i2v-14b-720p-Q3_K_S", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.1-i2v-14b-720p-Q3_K_S.gguf", - "size": 7925939584, - "modified": 1769096072.226664, - "sha256": "3a020d7b4c6e2e59135d8dd543e26af801217c73b3ac1d73bc9bf25420e8dc15", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": false, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": true, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 1769145840.493219, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.1-t2v-14b-Q3_K_S.gguf b/wan/wan2.1-t2v-14b-Q3_K_S.gguf deleted file mode 100644 index b6815e23e0d9751b3f00a2e4b3a524af9b42d732..0000000000000000000000000000000000000000 --- a/wan/wan2.1-t2v-14b-Q3_K_S.gguf +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8ba3d788a2f815b35469cf1920b2cd8c8ce7d90521442f90b15dfc3af8c54ab1 -size 6987864704 diff --git a/wan/wan2.1-t2v-14b-Q3_K_S.metadata.json b/wan/wan2.1-t2v-14b-Q3_K_S.metadata.json deleted file mode 100644 index 6e96491eb24d0a02b16019991a82d7fbef8ba2a5..0000000000000000000000000000000000000000 --- a/wan/wan2.1-t2v-14b-Q3_K_S.metadata.json +++ /dev/null @@ -1,23 +0,0 @@ -{ - "file_name": "wan2.1-t2v-14b-Q3_K_S", - "model_name": "wan2.1-t2v-14b-Q3_K_S", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.1-t2v-14b-Q3_K_S.gguf", - "size": 6987864704, - "modified": 1769096080.490471, - "sha256": "8ba3d788a2f815b35469cf1920b2cd8c8ce7d90521442f90b15dfc3af8c54ab1", - "base_model": "Unknown", - "preview_url": "", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": true, - "civitai": {}, - "tags": [], - "modelDescription": "", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": null, - "last_checked_at": 0, - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.metadata.json b/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.metadata.json deleted file mode 100644 index 40ddff429c0dd7a5b274c25bd897fefa1089a905..0000000000000000000000000000000000000000 --- a/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.metadata.json +++ /dev/null @@ -1,128 +0,0 @@ -{ - "file_name": "wan2.2_i2v_high_noise_14B_fp8_scaled", - "model_name": "Wan Video 2.2", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors", - "size": 14294742832, - "modified": 1769096141.462235, - "sha256": "6122e79d55e0f235698d11d657f3b196c5273c830da00b2b013c5a048d5e6a42", - "base_model": "Wan Video 2.2 I2V-A14B", - "preview_url": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.png", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2057465, - "modelId": 1817671, - "name": "i2v_high_noise_14B_fp8_sd", - "nsfwLevel": 1, - "createdAt": "2025-07-28T17:11:00.339Z", - "updatedAt": "2025-10-29T01:25:24.466Z", - "status": "Published", - "publishedAt": "2025-07-28T18:26:24.522Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Wan Video 2.2 I2V-A14B", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "wan2.2_i2v_high_noise_14B_fp8_scaled
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:wanvideo-22-i2v-a14b:checkpoint:civitai:1817671@2057465", - "stats": { - "downloadCount": 1577, - "thumbsUpCount": 65 - }, - "model": { - "name": "Wan Video 2.2", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "tags": [ - "base model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 1954246, - "sizeKB": 13959709.796875, - "name": "wanVideo22_i2vHighNoise14BFp8Sd.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-07-28T17:35:28.241Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "96651951", - "AutoV2": "6122E79D55", - "SHA256": "6122E79D55E0F235698D11D657F3B196C5273C830DA00B2B013C5A048D5E6A42", - "CRC32": "C9498879", - "BLAKE3": "88555708950BF6EACAB2CA3677832A6F953C8609E1F595C437F8634F9A1C0263", - "AutoV3": "E706C9293181" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2057465" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d05fadf4-0ed8-4a54-9c7d-9864ee36d31f/original=true/91022776.mp4", - "nsfwLevel": 1, - "width": 1280, - "height": 720, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 9193286, - "audio": false, - "width": 1280, - "height": 720, - "duration": 80.042 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2057465", - "creator": { - "username": "theally", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1e7869de-716f-4da9-80c4-91c770223210/width=96/theally.jpeg" - } - }, - "tags": [ - "base model" - ], - "modelDescription": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145851.077885, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.png b/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.png deleted file mode 100644 index 938a5b8c7b180b31031e248d28f86a220a9723f8..0000000000000000000000000000000000000000 --- a/wan/wan2.2_i2v_high_noise_14B_fp8_scaled.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f33a16fd16b23b108fd585b7ca9463987543a5685b46272c1b534bd8fc4a14de -size 9193286 diff --git a/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.metadata.json b/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.metadata.json deleted file mode 100644 index 8a423984500cd2d53bec789b14e0274116104c6a..0000000000000000000000000000000000000000 --- a/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.metadata.json +++ /dev/null @@ -1,128 +0,0 @@ -{ - "file_name": "wan2.2_i2v_low_noise_14B_fp8_scaled", - "model_name": "Wan Video 2.2", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors", - "size": 14294742832, - "modified": 1769096158.309155, - "sha256": "5471a457b6ac404202a5fbe6c11595a3d5641fc766b00f38763f72303fffc21e", - "base_model": "Wan Video 2.2 I2V-A14B", - "preview_url": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.png", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2057270, - "modelId": 1817671, - "name": "i2v_low_noise_14B_fp8_scd", - "nsfwLevel": 1, - "createdAt": "2025-07-28T16:09:45.883Z", - "updatedAt": "2025-10-29T01:25:00.187Z", - "status": "Published", - "publishedAt": "2025-07-28T17:09:47.098Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Wan Video 2.2 I2V-A14B", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "wan2.2_i2v_low_noise_14B_fp8_scaled
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:wanvideo-22-i2v-a14b:checkpoint:civitai:1817671@2057270", - "stats": { - "downloadCount": 1725, - "thumbsUpCount": 61 - }, - "model": { - "name": "Wan Video 2.2", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "tags": [ - "base model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 1954179, - "sizeKB": 13959709.796875, - "name": "wanVideo22_i2vLowNoise14BFp8Scd.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-07-28T17:13:19.942Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "C520652A", - "AutoV2": "5471A457B6", - "SHA256": "5471A457B6AC404202A5FBE6C11595A3D5641FC766B00F38763F72303FFFC21E", - "CRC32": "E4767827", - "BLAKE3": "8D25B11AA401E5D7E9F34A1B58DC4FF2473B182C3C37561A39ED8BBA5B0DAEA8", - "AutoV3": "AFBDE5842E01" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2057270" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c00e0424-2c09-4455-9c40-1ab04c44ec71/original=true/91012777.mp4", - "nsfwLevel": 1, - "width": 1280, - "height": 720, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 9193286, - "audio": false, - "width": 1280, - "height": 720, - "duration": 80.042 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2057270", - "creator": { - "username": "theally", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1e7869de-716f-4da9-80c4-91c770223210/width=96/theally.jpeg" - } - }, - "tags": [ - "base model" - ], - "modelDescription": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145851.298414, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.png b/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.png deleted file mode 100644 index 938a5b8c7b180b31031e248d28f86a220a9723f8..0000000000000000000000000000000000000000 --- a/wan/wan2.2_i2v_low_noise_14B_fp8_scaled.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f33a16fd16b23b108fd585b7ca9463987543a5685b46272c1b534bd8fc4a14de -size 9193286 diff --git a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.metadata.json b/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.metadata.json deleted file mode 100644 index d9bc6f1d27b7a3b831628b25dbacbcc9028275ec..0000000000000000000000000000000000000000 --- a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.metadata.json +++ /dev/null @@ -1,128 +0,0 @@ -{ - "file_name": "wan2.2_t2v_low_noise_14B_fp8_scaled", - "model_name": "Wan Video 2.2", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors", - "size": 14293923632, - "modified": 1769096174.092426, - "sha256": "e71b96d7c82e638694c5e7fb98fac4bfb0e4ddc5fbbb4b1df40da8f0f1278a97", - "base_model": "Wan Video 2.2 T2V-A14B", - "preview_url": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.mp4", - "preview_nsfw_level": 1, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2057100, - "modelId": 1817671, - "name": "t2v_low_noise_14B", - "nsfwLevel": 1, - "createdAt": "2025-07-28T15:23:17.855Z", - "updatedAt": "2025-10-29T01:24:02.075Z", - "status": "Published", - "publishedAt": "2025-07-28T15:42:32.840Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Wan Video 2.2 T2V-A14B", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "wan2.2_t2v_low_noise_14B_fp8_scaled
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:wanvideo-22-t2v-a14b:checkpoint:civitai:1817671@2057100", - "stats": { - "downloadCount": 1249, - "thumbsUpCount": 52 - }, - "model": { - "name": "Wan Video 2.2", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "tags": [ - "base model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 1953915, - "sizeKB": 13958909.796875, - "name": "wanVideo22_t2vLowNoise14B.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-07-28T15:53:05.451Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "053A1060", - "AutoV2": "E71B96D7C8", - "SHA256": "E71B96D7C82E638694C5E7FB98FAC4BFB0E4DDC5FBBB4B1DF40DA8F0F1278A97", - "CRC32": "2E14F608", - "BLAKE3": "C91B6776BD509549C533CD3A99B47EBF5F8728F5616A3DA26B605C3317793F9C", - "AutoV3": "698EB11168F4" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2057100" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0cf9f668-561b-4c50-b86d-fb47b635f019/original=true/91000464.mp4", - "nsfwLevel": 1, - "width": 1280, - "height": 720, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 9193286, - "audio": false, - "width": 1280, - "height": 720, - "duration": 80.042 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2057100", - "creator": { - "username": "theally", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1e7869de-716f-4da9-80c4-91c770223210/width=96/theally.jpeg" - } - }, - "tags": [ - "base model" - ], - "modelDescription": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145851.481418, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.mp4 b/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.mp4 deleted file mode 100644 index 86b0ffec00d3018ca0a1719e479ee7060b865f68..0000000000000000000000000000000000000000 --- a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.mp4 +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ee879367db7ea8a5728a22730cdf9ef0db2ebdd53082680111d68bc37b24d2c1 -size 9892534 diff --git a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.png b/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.png deleted file mode 100644 index 938a5b8c7b180b31031e248d28f86a220a9723f8..0000000000000000000000000000000000000000 --- a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f33a16fd16b23b108fd585b7ca9463987543a5685b46272c1b534bd8fc4a14de -size 9193286 diff --git a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors b/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors deleted file mode 100644 index 71754b5971c85324cf09684d08f199a6ffa8302a..0000000000000000000000000000000000000000 --- a/wan/wan2.2_t2v_low_noise_14B_fp8_scaled.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e71b96d7c82e638694c5e7fb98fac4bfb0e4ddc5fbbb4b1df40da8f0f1278a97 -size 14293923632 diff --git a/wan/wan2.2_ti2v_5B_fp16.metadata.json b/wan/wan2.2_ti2v_5B_fp16.metadata.json deleted file mode 100644 index f49cb13fb002977ccae163e729e9ef21728862f6..0000000000000000000000000000000000000000 --- a/wan/wan2.2_ti2v_5B_fp16.metadata.json +++ /dev/null @@ -1,180 +0,0 @@ -{ - "file_name": "wan2.2_ti2v_5B_fp16", - "model_name": "Wan Video 2.2", - "file_path": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_ti2v_5B_fp16.safetensors", - "size": 9999658848, - "modified": 1769096185.275691, - "sha256": "456f901338bd9eadbded3828b819109a9b68e8a525ca5cf8d0049a69fcfeca1e", - "base_model": "Wan Video 2.2 TI2V-5B", - "preview_url": "C:/ComfyUI/models/diffusion_models/wan/wan2.2_ti2v_5B_fp16.png", - "preview_nsfw_level": 0, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2057016, - "modelId": 1817671, - "name": "ti2v_5B_fp16", - "nsfwLevel": 1, - "createdAt": "2025-07-28T14:54:51.559Z", - "updatedAt": "2025-10-29T01:22:53.231Z", - "status": "Published", - "publishedAt": "2025-07-28T15:13:30.403Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "Wan Video 2.2 TI2V-5B", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "wan2.2_ti2v_5B_fp16
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:wanvideo-22-ti2v-5b:checkpoint:civitai:1817671@2057016", - "stats": { - "downloadCount": 2371, - "thumbsUpCount": 98 - }, - "model": { - "name": "Wan Video 2.2", - "type": "Checkpoint", - "nsfw": false, - "poi": false, - "description": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "tags": [ - "base model" - ], - "allowNoCredit": true, - "allowCommercialUse": [ - "Image", - "RentCivit", - "Rent", - "Sell" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 1953835, - "sizeKB": 9765291.84375, - "name": "wanVideo22_ti2v5BFp16.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-07-28T15:19:31.065Z", - "metadata": { - "format": "SafeTensor", - "size": "full", - "fp": "fp16" - }, - "hashes": { - "AutoV1": "D6B99589", - "AutoV2": "456F901338", - "SHA256": "456F901338BD9EADBDED3828B819109A9B68E8A525CA5CF8D0049A69FCFECA1E", - "CRC32": "CB351628", - "BLAKE3": "322880D094897C98ED02FA4D264A1CE254B3239AC4EB81A7172BEB3E4F7FA397", - "AutoV3": "77FB849EDE60" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2057016" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/10ad5006-1fdd-4652-be9b-a727af9c274b/original=true/91190530.mp4", - "nsfwLevel": 1, - "width": 704, - "height": 1280, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 2392238, - "audio": false, - "width": 704, - "height": 1280, - "duration": 5.042 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "a beautiful woman, subtle movements, posing, laughing" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a7e4ce02-481b-4b96-80a3-906b25dd86a5/original=true/91195367.mp4", - "nsfwLevel": 1, - "width": 704, - "height": 1280, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 869212, - "audio": false, - "width": 704, - "height": 1280, - "duration": 5.042 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "a handsome victorian man, talking and smiling" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b68d87b5-b3c2-4805-8d40-54c8a518b190/original=true/90996924.mp4", - "nsfwLevel": 1, - "width": 1280, - "height": 720, - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "type": "video", - "metadata": { - "hash": "U00000fQfQfQfQfQfQfQfQfQfQfQfQfQfQfQ", - "size": 9193286, - "audio": false, - "width": 1280, - "height": 720, - "duration": 80.042 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2057016", - "creator": { - "username": "theally", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1e7869de-716f-4da9-80c4-91c770223210/width=96/theally.jpeg" - } - }, - "tags": [ - "base model" - ], - "modelDescription": "Note: There are other Wan Video files hosted on Civitai - these may be duplicates, but this model card is primarily to host the files used by Wan Video in the Civitai Generator.
These files are the ComfyUI Repack - the original files can be found in Diffusers/multi-part safetensors format here.
Wan2.2, a major upgrade to our visual generative models, which is now open-sourced, offering more powerful capabilities, better performance, and superior visual quality. With Wan2.2, we have focused on incorporating the following technical innovations:
👍 MoE Architecture: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost.
💪🏻 Data Scaling: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models.
🎬 Cinematic Aesthetics: Wan2.2 incorporates specially curated aesthetic data with fine-grained labels for lighting, composition, and color. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences.
🚀 Efficient High-Definition Hybrid TI2V: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It one of the fastest 720P@24fps models currently available, capable of serving both the industrial and academic sectors simultaneously.
The T2V-A14B model, supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On our new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across most key evaluation dimensions.
The I2V-A14B model, designed for image-to-video generation, supports both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for diverse stylized scenes.
The TI2V-5B model is built with the advanced Wan2.2-VAE that achieves a compression ratio of 16×16×4. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can runs on single consumer-grade GPU such as the 4090. It is one of the fastest 720P@24fps models available, meeting the needs of both industrial applications and academic research.
GitHub: https://github.com/Wan-Video/Wan2.2
Originally HuggingFace Repo: https://huggingface.co/Comfy-Org/Wan_2.2_ComfyUI_Repackaged/tree/main/split_files/diffusion_models
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145852.176887, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/wan/wan2.2_ti2v_5B_fp16.png b/wan/wan2.2_ti2v_5B_fp16.png deleted file mode 100644 index 43cbd69fff2d910c93c559d35ff41974a67a74a2..0000000000000000000000000000000000000000 --- a/wan/wan2.2_ti2v_5B_fp16.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d01eba975f4811e52633f98247b0b9a74449343e3fe0aac5737aaa4ae89c4b54 -size 2392238 diff --git a/zimageturbo/moodyPornMix_zitV3.jpeg b/zimageturbo/moodyPornMix_zitV3.jpeg deleted file mode 100644 index 8f51fd934c60509e457d8ae38145482265f88089..0000000000000000000000000000000000000000 Binary files a/zimageturbo/moodyPornMix_zitV3.jpeg and /dev/null differ diff --git a/zimageturbo/moodyPornMix_zitV3.metadata.json b/zimageturbo/moodyPornMix_zitV3.metadata.json deleted file mode 100644 index 02852151401301fe6e9f5fe7147864e96e349440..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV3.metadata.json +++ /dev/null @@ -1,364 +0,0 @@ -{ - "file_name": "moodyPornMix_zitV3", - "model_name": "Moody Porn Mix", - "file_path": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV3.safetensors", - "size": 12309887144, - "modified": 1769096197.942555, - "sha256": "3ae639667b151566f860c1445e1d20cdbcc56ba287abbf8cf1e957548156b6f0", - "base_model": "ZImageTurbo", - "preview_url": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV3.jpeg", - "preview_nsfw_level": 4, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2533797, - "modelId": 620406, - "name": "ZIT V3", - "nsfwLevel": 60, - "createdAt": "2025-12-25T01:53:03.838Z", - "updatedAt": "2026-01-16T13:03:45.933Z", - "status": "Published", - "publishedAt": "2025-12-25T02:06:46.835Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "ZImageTurbo", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": "MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also up and running.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
", - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:zimageturbo:checkpoint:civitai:620406@2533797", - "stats": { - "downloadCount": 4332, - "thumbsUpCount": 307 - }, - "model": { - "name": "Moody Porn Mix", - "type": "Checkpoint", - "nsfw": true, - "poi": false, - "description": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "allowNoCredit": false, - "allowCommercialUse": [ - "RentCivit", - "Rent" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2421663, - "sizeKB": 12021374.1640625, - "name": "moodyPornMix_zitV3.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-12-25T02:04:12.162Z", - "metadata": { - "format": "SafeTensor", - "size": "full", - "fp": "fp16" - }, - "hashes": { - "AutoV1": "D21E9777", - "AutoV2": "3AE639667B", - "SHA256": "3AE639667B151566F860C1445E1D20CDBCC56BA287ABBF8CF1E957548156B6F0", - "CRC32": "215F3099", - "BLAKE3": "81624EB66A162A4C995DFBB3866BFF058BB94449AB15B347135EE6B7B937B1D0", - "AutoV3": "BFE34B719A5B" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2533797" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ff321e56-eb40-4420-97f9-92b52e27dde4/original=true/115064385.jpeg", - "nsfwLevel": 4, - "width": 2224, - "height": 3344, - "hash": "UNJ*F-D%%g%M.TaJkWx]JB%gxZRPIAx]oIIU", - "type": "image", - "metadata": { - "hash": "UNJ*F-D%%g%M.TaJkWx]JB%gxZRPIAx]oIIU", - "size": 8851702, - "width": 2224, - "height": 3344 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一个清纯可爱的16岁日本少女，安静地蜷缩在柔软的米白色绒面沙发上，穿着浅粉色棉质超大落肩毛衣（oversized dropshoulder sweater），宽松慵懒的落肩设计使肩线自然下垂至上臂中段，袖子格外宽大而修长，从肩膀完全滑落，堆叠出柔软的褶皱层层覆盖手臂，只露出纤细白皙的手腕和指尖，圆领口松弛低垂，柔软轻薄的棉织物顺着身体曲线自然敞开，形成诱人而深邃的V形乳沟，乳沟阴影柔和渐变，中央肌肤白皙细腻饱满，边缘被领口布料轻轻包裹，若隐若现地勾勒出少女圆润挺拔的胸部轮廓与自然弧线，衣摆宽松延伸至大腿中上部，布料蓬松堆积出温暖的层层叠叠褶皱，下摆微微卷边，触感柔软亲肤，整体毛衣材质细腻保暖，带着淡淡棉香与少女体温；修长白皙的双腿随意叠起，膝盖轻轻弯曲，黑色直发柔顺披肩，发尾微微卷翘，丝滑光泽，头戴银色AirPods Max耳机，耳罩轻压发丝，精致五官如瓷娃娃般纯净，大眼睛专注温柔凝视手中厚厚圣经，书封金色浮雕“HOLY BIBLE”，粉嫩脸颊被午后阳光映照泛起娇羞红晕，嘴角恬静浅笑，眼角微微上翘，散发纯真平和；双手轻捧书本，指尖纤细白皙，书页泛黄，沙发旁小圆桌热腾腾红茶蒸汽袅袅，杯旁精致饼干，背景落地窗外冰雪世界雪花飘落覆盖松树高楼，室内暖光与窗外冷白柔和对比，金色阳光斜洒穿过薄纱窗帘，在少女脸庞、锁骨、深邃乳沟与落肩毛衣褶皱上投下斑驳光影与温暖金色光晕，细腻勾勒皮肤纹理、乳沟自然阴影曲线与布料质感，浅景深摄影风格，焦点锐利集中于少女脸庞、乳沟细节、毛衣领口褶皱、书本与双手，背景雪景轻微虚化，整体色调温暖柔和偏粉金，充满宁静治愈冬日温馨高品质氛围，高度真实细腻皮肤质感、自然光晕毛孔细节，少女肌肤白皙透亮如雪，乳沟处光滑饱满，散发少女独有清新纯真与诱人美感，高光柔和反射于毛衣布料、头发与耳机，增强立体感，低角度视角从沙发边缘向上拍摄，突出少女柔美身材曲线与亲切感，画面精致优雅如高端杂志封面。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8f55c381-8da3-427a-96a3-99db0bbbcf5c/original=true/115064550.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UCEMF5Xn03%1~qaeI@ozOmNHe9RPV;wdIToe", - "type": "image", - "metadata": { - "hash": "UCEMF5Xn03%1~qaeI@ozOmNHe9RPV;wdIToe", - "size": 14221873, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "在2077年的未来赛博朋克世界中，一张全景全身照捕捉到一个19岁的年轻日本女性，她拥有长长的乌黑直发和齐整的刘海，精致的五官带着一丝迷茫而诱人的表情，樱桃小嘴微微张开，眼睛直视镜头透露出好奇与无奈。她被困在一个巨大的透明玻璃容器中，这个容器放置在高科技画廊的展示台上，容器表面贴着醒目的标签“2025 Human Specimen”, \"Female Variant\"，周围环绕着闪烁的霓虹灯光和 holographic 显示屏，营造出科幻而压抑的氛围。她完全裸体，露出华丽丰满的身材曲线，巨大的自然下垂乳房丰盈而真实，大圆臀部光滑诱人，身上散布着几颗细小的痣，皮肤纹理清晰细腻，宛如瓷器般光滑。她站立面对镜头，一手优雅地搁在腰间，另一手按在玻璃上，手指涂抹着可爱粉红色的指甲油，姿势自信而妩媚，脚部清晰可见，脚趾修长匀称。构图采用正面视角，焦点集中在她的身躯上，背景中融合了赛博朋克元素如电路板纹路、荧光蓝紫色调和金属反射。光线柔和却带有未来感，温暖的金色光从上方洒下，形成柔美的阴影和高光对比，色彩风格以冷峻的蓝绿色为主调，点缀鲜艳的霓虹粉红，整体呈现高品质的美学，杰作级细节，8K超高清画质，皮肤纹理逼真，动态光影效果增强了体积感和深度感。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0022a534-77b1-4e19-a6ee-8a0d2ad17f6b/original=true/115064542.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "U3F6IdGRD].9Tw#q9[+F009@?a9[8wvi}.TL", - "type": "image", - "metadata": { - "hash": "U3F6IdGRD].9Tw#q9[+F009@?a9[8wvi}.TL", - "size": 14539568, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一个清纯却大胆的俄罗斯女coser坐在科幻主题展厅中央的透明悬浮座椅上，双腿张开跨坐在透明水晶凳上，甜美纯净的娃娃脸带着羞涩却自信的微笑，丰满巨乳完全裸露在外，粉嫩乳头与大片乳晕清晰可见，纤细腰肢与爆炸性胸臀曲线形成极致沙漏身材，她穿着超短的银白色未来科技风格魔法少女装，透明塑料与金属光泽荷叶边裙摆几乎不遮下体，完全无内裤，随着双腿自信分开，裙下光洁无毛的私处毫无保留地暴露，右手握着一把发光蓝色能量魔法杖随意垂在身侧，脚踩银色透明高跟靴衬托出修长笔直小腿，中景低角度仰拍强调她大胆暴露的坐姿与未来感色情魔法少女装扮，背景是科幻展厅、闪烁的全息投影与霓虹科技摊位，冷蓝顶光结合紫色与青色氛围灯形成强烈科技反差，锐利焦点落在胸部与下体细节，超高精度科幻摄影风格，细腻皮肤纹理、真实金属与塑料质感，皮肤白皙如雪，曲线在冷光下泛着冰冷又性感的光泽，充满未来感与挑逗的露出氛围。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b2486a05-6072-47bb-8266-c4c0a76f2a86/original=true/115064543.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UFHA;HEzNHR*~CNdNGRjR$NaRQRj-noLR+xa", - "type": "image", - "metadata": { - "hash": "UFHA;HEzNHR*~CNdNGRjR$NaRQRj-noLR+xa", - "size": 14907013, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一个妖娆丰满的中国女coser坐在华丽的动漫咖啡厅中央的玻璃圆桌上，双腿张开跨坐在透明水晶凳上，精致妩媚的鹅蛋脸带着挑逗的甜笑，巨乳傲然挺立完全裸露，粉红乳头与大片乳晕一览无余，盈盈一握的细腰与夸张的丰臀形成强烈S型曲线，她穿着极短的酒红色中式旗袍式魔法少女装，绸缎面料高开叉搭配层层薄纱荷叶边，裙摆短到无法遮挡下体，完全无内裤，随着双腿自信分开，裙下光滑无毛的私处完全暴露，右手优雅握着一把古典团扇魔法道具垂在身侧，脚踩红色绸缎高跟鞋勾勒出纤细脚踝与修长小腿，中景低角度仰拍突出她暴露而优雅的坐姿与中式色情魔法少女装扮，背景是温馨的咖啡厅、穿着cos服的客人与彩色吊灯，暖黄顶光结合粉红与金色氛围灯营造柔美反差，锐利焦点集中于胸部与下体细节，电影级真实摄影风格，超细腻皮肤纹理、真实绸缎反光与毛孔质感，皮肤白皙透亮如瓷，油光发亮，曲线在暖光下泛着诱人光泽，性感而高雅的露出氛围。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e36dd36b-b9cb-4029-a01f-e474ce2102d2/original=true/115064548.jpeg", - "nsfwLevel": 8, - "width": 2608, - "height": 3880, - "hash": "ULKJP?w$Ip%2~ExbWBoMtSxuIURkMws,NHR*", - "type": "image", - "metadata": { - "hash": "ULKJP?w$Ip%2~ExbWBoMtSxuIURkMws,NHR*", - "size": 11389805, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一个现代化房间内，墙壁上闪烁着醒目的“SEX PET ADOPTION”霓虹灯，空间干净而弥漫禁忌压抑氛围，地面光滑浅色大理石瓷砖；内部关押两位全裸年轻女性，如被囚禁的性感宠物般暴露展示，地面散落鲜艳狗玩具包括红色橡胶球、骨头啃咬玩具、毛绒小狗玩偶和小狗垫子，以及闪亮不锈钢狗水盆，强化禁忌宠物化情境；\n\n前景金发美女皮肤白皙丰满，19岁模样，全裸正面面向镜头，脖子带着狗环，低蹲姿势双腿大胆张开完全露出私密部位，臀部翘起，双手向前伸出握成拳头放在胸前，像狗狗的前爪，头部微微前倾，胸部挺起，面带挑逗媚笑，舌尖轻舔唇角，眼神饥渴俏皮直视观众，长金发凌乱散落，肌肤细腻湿润泛汗珠，乳头与私处细节清晰，姿态充满强烈情色张力；\n\n背景角落瘦小女性，25岁模样，日本人，脖子带着狗环，胸部扁平，全裸侧坐发呆，表情茫然顺从，双腿蜷曲靠墙，身体略模糊增强纵深；\n\n广角低角度仰视构图，营造淫靡压抑梦幻深度；照明柔和粉紫暖橙情调灯，顶部灯带散射暧昧柔光，隔间内暖黄聚光突出肌肤湿润高光与深邃阴影；色调温暖粉红桃红金橙为主，点缀深紫阴影与诱人肌肤光泽，热烈超现实色情氛围；高度写实情色风格，当代艺术摄影大师级，细腻纹理情感张力，超现实美学，8K超高清，电影级光影渲染。\n" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b3d65787-b602-4614-958f-2eb30f8eefd7/original=true/115064516.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UCJQyU8_%Nt6_NE1~WM|.9rr-pn*NdRkO@M{", - "type": "image", - "metadata": { - "hash": "UCJQyU8_%Nt6_NE1~WM|.9rr-pn*NdRkO@M{", - "size": 14140888, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一位身材高挑、模特般修长的中国女大学生，约20岁，皮肤白皙细腻如瓷器，仰卧在凌乱的女生宿舍单人床上，床单皱巴巴，散落着衣物、书籍和零食。她正面面向镜头，双腿高高举起并大大分开成V字形，膝盖微微弯曲，精致玉足向上，脚趾自然蜷曲，脚底粉嫩。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/dfb43c76-b130-4906-aae4-e7f38def4c32/original=true/115064969.jpeg", - "nsfwLevel": 8, - "width": 2608, - "height": 3880, - "hash": "UFIzk-x[AbV@Z#-UafXT0Ji]IoM{~Wf+IpRP", - "type": "image", - "metadata": { - "hash": "UFIzk-x[AbV@Z#-UafXT0Ji]IoM{~Wf+IpRP", - "size": 11050019, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "\n一个18岁可爱台湾女孩和一个25岁韩国辣妹，两人上身完全赤裸，面对镜头。韩国辣妹站在前景（镜头最前面，占据画面主要位置），台湾女孩跪在后面床上（明显处于后排位置，形成前后层次）。\n\n韩国辣妹（前景，站立在地毯上）：\n25岁，干瘦高挑骨感身材，冷艳尖下巴、高鼻梁的骨感脸庞，妆容精致性感，烟熏眼妆凌厉，薄唇上扬露出夸张的嘲讽冷笑，表情强势狰狞却带着搞笑扭曲的愤怒感，长直黑发披散纠缠，单手叉腰，另一只手向后指着台湾妹妹，十分生气的样子，充满霸气与戏剧性张力。\n\n台湾女孩（后景，跪在床上）：\n18岁可爱台湾妹妹，身材娇小矮小，齐刘海齐肩微卷短发凌乱飞扬，苹果般圆脸、大眼睛水汪汪、婴儿肥脸颊泛红，粉嫩樱桃小嘴撅起满是委屈和不甘，低着头看向地面，双手放在身前，丰满胸部因拉扯动作剧烈晃动，表情楚楚可怜。\n\n场景：\n\n浪漫却凌乱的酒店房间，床单凌乱、衣物散落，空气中弥漫粉色暧昧氛围。暖色灯光柔和勾勒两人肌肤的高光与细腻纹理，细节丰富，高品质写实风格，融合强烈色情诱惑与幽默搞笑元素，画面生动性感又充满戏剧性张力，美轮美奂。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/8e0dfaee-4d21-4e91-8bf7-a538c6ff2f65/original=true/115064976.jpeg", - "nsfwLevel": 16, - "width": 2608, - "height": 3880, - "hash": "U5I|wSIAK*9E01?bM_M|15M{?H^j~Xb_rqEM", - "type": "image", - "metadata": { - "hash": "U5I|wSIAK*9E01?bM_M|15M{?H^j~Xb_rqEM", - "size": 11341610, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一个18岁可爱台湾女孩和一个25岁韩国辣妹在卧室里激烈争抢同一个男人，两人上身完全赤裸，露出腿和臀部曲线；" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/7c8fc6e0-3cba-445f-992d-836e69dd78ae/original=true/115070326.jpeg", - "nsfwLevel": 8, - "width": 3936, - "height": 5912, - "hash": "UEEx@dI:01xv~qniIUNbOqsAVrR*-oWXE0%1", - "type": "image", - "metadata": { - "hash": "UEEx@dI:01xv~qniIUNbOqsAVrR*-oWXE0%1", - "size": 24069113, - "width": 3936, - "height": 5912 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "在2077年的未来赛博朋克世界中，一张全景全身照捕捉到一个19岁的年轻日本女性，她拥有长长的乌黑直发和齐整的刘海，精致的五官带着一丝迷茫而诱人的表情，樱桃小嘴微微张开，眼睛直视镜头透露出好奇与无奈。她被困在一个巨大的透明玻璃容器中，这个容器放置在高科技画廊的展示台上，容器表面贴着醒目的标签“2060 Human Specimen”, \"Female Variant\", \"FAT\"，周围环绕着闪烁的霓虹灯光和 holographic 显示屏，营造出科幻而压抑的氛围。她完全裸体，露出华丽丰满的身材曲线，巨大的自然下垂乳房丰盈而真实，Saggy breasts, 大圆臀部光滑诱人，身上散布着几颗细小的痣，皮肤纹理清晰细腻，宛如瓷器般光滑。她站立面对镜头，另两只手都比V字，手指涂抹着可爱粉蓝色的指甲油，姿势自信而妩媚，脚部清晰可见，脚趾修长匀称。构图采用正面视角，焦点集中在她的身躯上，背景中融合了赛博朋克元素如电路板纹路、荧光蓝紫色调和金属反射。光线柔和却带有未来感，温暖的金色光从上方洒下，形成柔美的阴影和高光对比，色彩风格以冷峻的蓝绿色为主调，点缀鲜艳的霓虹粉红，整体呈现高品质的美学，杰作级细节，8K超高清画质，皮肤纹理逼真，动态光影效果增强了体积感和深度感。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c3c108c7-f6fc-4c07-91ab-975151b682c4/original=true/115070344.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "ULEL{aIo0g$~~WR*IpjZJ6WBrqX8t2o3Ins:", - "type": "image", - "metadata": { - "hash": "ULEL{aIo0g$~~WR*IpjZJ6WBrqX8t2o3Ins:", - "size": 15070950, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "在2077年的未来赛博朋克世界中，一张全景全身照捕捉到一个19岁的年轻日本女性，她拥有长长的乌黑直发和齐整的刘海，精致的五官带着一丝迷茫而诱人的表情，樱桃小嘴微微张开，眼睛直视镜头透露出好奇与无奈。她被困在一个巨大的透明玻璃容器中，这个容器放置在高科技画廊的展示台上，容器表面贴着醒目的标签“2060 Human Specimen”, \"Female Variant\", \"FAT\"，周围环绕着闪烁的霓虹灯光和 holographic 显示屏，营造出科幻而压抑的氛围。她完全裸体，露出华丽丰满的身材曲线，巨大的自然下垂乳房丰盈而真实，Saggy breasts, 大圆臀部光滑诱人，身上散布着几颗细小的痣，皮肤纹理清晰细腻，宛如瓷器般光滑。她站立面对镜头，另两只手都比V字，手指涂抹着可爱粉蓝色的指甲油，姿势自信而妩媚，脚部清晰可见，脚趾修长匀称。构图采用正面视角，焦点集中在她的身躯上，背景中融合了赛博朋克元素如电路板纹路、荧光蓝紫色调和金属反射。光线柔和却带有未来感，温暖的金色光从上方洒下，形成柔美的阴影和高光对比，色彩风格以冷峻的蓝绿色为主调，点缀鲜艳的霓虹粉红，整体呈现高品质的美学，杰作级细节，8K超高清画质，皮肤纹理逼真，动态光影效果增强了体积感和深度感。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2533797", - "creator": { - "username": "catlover1937", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/18485719-777f-43e9-b4ae-7ae3000e5468/width=96/catlover1937.jpeg" - } - }, - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "modelDescription": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145852.489155, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/zimageturbo/moodyPornMix_zitV3.png b/zimageturbo/moodyPornMix_zitV3.png deleted file mode 100644 index 19b52bc1b66b3b82401cf847f3d67deead7c6745..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV3.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9f40778d76488ce08baba9951c88730e73a73c72cfcef477d6dadc5cbbe11e29 -size 8851702 diff --git a/zimageturbo/moodyPornMix_zitV3FP8.jpeg b/zimageturbo/moodyPornMix_zitV3FP8.jpeg deleted file mode 100644 index 12d1ccd1d5e9040db308019d0918f14f8ca7dd68..0000000000000000000000000000000000000000 Binary files a/zimageturbo/moodyPornMix_zitV3FP8.jpeg and /dev/null differ diff --git a/zimageturbo/moodyPornMix_zitV3FP8.metadata.json b/zimageturbo/moodyPornMix_zitV3FP8.metadata.json deleted file mode 100644 index cdeef2db7a0404348a30f92edcd36e62cdece941..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV3FP8.metadata.json +++ /dev/null @@ -1,300 +0,0 @@ -{ - "file_name": "moodyPornMix_zitV3FP8", - "model_name": "Moody Porn Mix", - "file_path": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV3FP8.safetensors", - "size": 6154975672, - "modified": 1769096204.665837, - "sha256": "378b68c9a9067d3b3b9c4c0264825a0c7559c2ebcfaa86df8663f012fe9cacf0", - "base_model": "ZImageTurbo", - "preview_url": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV3FP8.jpeg", - "preview_nsfw_level": 8, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2533850, - "modelId": 620406, - "name": "ZIT V3 FP8", - "nsfwLevel": 60, - "createdAt": "2025-12-25T02:18:42.406Z", - "updatedAt": "2026-01-16T13:03:45.933Z", - "status": "Published", - "publishedAt": "2025-12-25T02:21:54.872Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "ZImageTurbo", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": null, - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:zimageturbo:checkpoint:civitai:620406@2533850", - "stats": { - "downloadCount": 1260, - "thumbsUpCount": 41 - }, - "model": { - "name": "Moody Porn Mix", - "type": "Checkpoint", - "nsfw": true, - "poi": false, - "description": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "allowNoCredit": false, - "allowCommercialUse": [ - "RentCivit", - "Rent" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2421699, - "sizeKB": 6010718.4296875, - "name": "moodyPornMix_zitV3FP8.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2025-12-25T02:27:17.166Z", - "metadata": { - "format": "SafeTensor", - "size": "pruned", - "fp": "fp8" - }, - "hashes": { - "AutoV1": "8A4AD4AC", - "AutoV2": "378B68C9A9", - "SHA256": "378B68C9A9067D3B3B9C4C0264825A0C7559C2EBCFAA86DF8663F012FE9CACF0", - "CRC32": "66AF9928", - "BLAKE3": "460E8161E45216AABCF7D54B4038C19A7BF19C3B0EB9EC13DDD98A7843EA1049", - "AutoV3": "77A51000258C" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2533850" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/73f4e116-e2f7-4f21-bdb6-b034e7d8cc7f/original=true/115066532.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UCEMF5Xn03%1~qaeI@ozOmNHe9RPV;wdIToe", - "type": "image", - "metadata": { - "hash": "UCEMF5Xn03%1~qaeI@ozOmNHe9RPV;wdIToe", - "size": 14221873, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e04ee3fd-eba1-436b-9b35-25746e460557/original=true/115066528.jpeg", - "nsfwLevel": 4, - "width": 2224, - "height": 3344, - "hash": "UNJ*F-D%%g%M.TaJkWx]JB%gxZRPIAx]oIIU", - "type": "image", - "metadata": { - "hash": "UNJ*F-D%%g%M.TaJkWx]JB%gxZRPIAx]oIIU", - "size": 8851702, - "width": 2224, - "height": 3344 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/79326193-1934-4d41-93f7-100098030a22/original=true/115066533.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UCJQyU8_%Nt6_NE1~WM|.9rr-pn*NdRkO@M{", - "type": "image", - "metadata": { - "hash": "UCJQyU8_%Nt6_NE1~WM|.9rr-pn*NdRkO@M{", - "size": 14140888, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1559e1a8-bf98-43c9-a6bc-5b09312bc7bc/original=true/115066535.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "U3F6IdGRD].9Tw#q9[+F009@?a9[8wvi}.TL", - "type": "image", - "metadata": { - "hash": "U3F6IdGRD].9Tw#q9[+F009@?a9[8wvi}.TL", - "size": 14539568, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a5f65c66-552c-416a-a67c-dc16264d2980/original=true/115066534.jpeg", - "nsfwLevel": 8, - "width": 2960, - "height": 4448, - "hash": "UFHA;HEzNHR*~CNdNGRjR$NaRQRj-noLR+xa", - "type": "image", - "metadata": { - "hash": "UFHA;HEzNHR*~CNdNGRjR$NaRQRj-noLR+xa", - "size": 14907013, - "width": 2960, - "height": 4448 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/35fc4aba-05c7-4679-a5f3-f5dc3bb885d7/original=true/115066530.jpeg", - "nsfwLevel": 8, - "width": 2608, - "height": 3880, - "hash": "UFIzk-x[AbV@Z#-UafXT0Ji]IoM{~Wf+IpRP", - "type": "image", - "metadata": { - "hash": "UFIzk-x[AbV@Z#-UafXT0Ji]IoM{~Wf+IpRP", - "size": 11050019, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/140ec91a-cf32-4eab-ad65-8f1785a37cdd/original=true/115066531.jpeg", - "nsfwLevel": 16, - "width": 2608, - "height": 3880, - "hash": "U5I|wSIAK*9E01?bM_M|15M{?H^j~Xb_rqEM", - "type": "image", - "metadata": { - "hash": "U5I|wSIAK*9E01?bM_M|15M{?H^j~Xb_rqEM", - "size": 11341610, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ede67cf3-c4f6-4fdc-9de9-3692755322a5/original=true/115066529.jpeg", - "nsfwLevel": 8, - "width": 2608, - "height": 3880, - "hash": "ULKJP?w$Ip%2~ExbWBoMtSxuIURkMws,NHR*", - "type": "image", - "metadata": { - "hash": "ULKJP?w$Ip%2~ExbWBoMtSxuIURkMws,NHR*", - "size": 11389805, - "width": 2608, - "height": 3880 - }, - "minor": false, - "poi": false, - "meta": null, - "availability": "Public", - "hasMeta": false, - "hasPositivePrompt": false, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2533850", - "creator": { - "username": "catlover1937", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/18485719-777f-43e9-b4ae-7ae3000e5468/width=96/catlover1937.jpeg" - } - }, - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "modelDescription": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145852.763061, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/zimageturbo/moodyPornMix_zitV3FP8.png b/zimageturbo/moodyPornMix_zitV3FP8.png deleted file mode 100644 index 19045fe9db0aa4cee99ce9746b15b41eae6203a0..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV3FP8.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:923d35091ac53fc5906517325aee8f6b855c560a336169456faa76b1a9f1375a -size 14221873 diff --git a/zimageturbo/moodyPornMix_zitV6.jpeg b/zimageturbo/moodyPornMix_zitV6.jpeg deleted file mode 100644 index 41e5210245c56d17086e21f5ab48b8bd01a66bbd..0000000000000000000000000000000000000000 Binary files a/zimageturbo/moodyPornMix_zitV6.jpeg and /dev/null differ diff --git a/zimageturbo/moodyPornMix_zitV6.metadata.json b/zimageturbo/moodyPornMix_zitV6.metadata.json deleted file mode 100644 index a6a494858462eda59aa93c5ac29d798ba6b09614..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV6.metadata.json +++ /dev/null @@ -1,364 +0,0 @@ -{ - "file_name": "moodyPornMix_zitV6", - "model_name": "Moody Porn Mix", - "file_path": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV6.safetensors", - "size": 12309879344, - "modified": 1769096218.044116, - "sha256": "8f7156b1a9993d4bc53d02e6a55bbfacda17bae64cf2df2bbb8582dc60d7533a", - "base_model": "ZImageTurbo", - "preview_url": "C:/ComfyUI/models/diffusion_models/zimageturbo/moodyPornMix_zitV6.jpeg", - "preview_nsfw_level": 8, - "notes": "", - "from_civitai": true, - "civitai": { - "id": 2602723, - "modelId": 620406, - "name": "ZIT-V6", - "nsfwLevel": 60, - "createdAt": "2026-01-16T12:05:12.154Z", - "updatedAt": "2026-01-16T13:03:45.933Z", - "status": "Published", - "publishedAt": "2026-01-16T13:03:37.432Z", - "trainedWords": [], - "trainingStatus": null, - "trainingDetails": null, - "baseModel": "ZImageTurbo", - "baseModelType": "Standard", - "earlyAccessEndsAt": null, - "earlyAccessConfig": null, - "description": null, - "uploadType": "Created", - "usageControl": "Download", - "air": "urn:air:zimageturbo:checkpoint:civitai:620406@2602723", - "stats": { - "downloadCount": 5647, - "thumbsUpCount": 403 - }, - "model": { - "name": "Moody Porn Mix", - "type": "Checkpoint", - "nsfw": true, - "poi": false, - "description": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "allowNoCredit": false, - "allowCommercialUse": [ - "RentCivit", - "Rent" - ], - "allowDerivatives": true, - "allowDifferentLicense": true - }, - "files": [ - { - "id": 2490019, - "sizeKB": 12021366.546875, - "name": "moodyPornMix_zitV6.safetensors", - "type": "Model", - "pickleScanResult": "Success", - "pickleScanMessage": "No Pickle imports", - "virusScanResult": "Success", - "virusScanMessage": null, - "scannedAt": "2026-01-16T12:19:26.179Z", - "metadata": { - "format": "SafeTensor", - "size": "full", - "fp": "fp16" - }, - "hashes": { - "AutoV1": "395B37B9", - "AutoV2": "8F7156B1A9", - "SHA256": "8F7156B1A9993D4BC53D02E6A55BBFACDA17BAE64CF2DF2BBB8582DC60D7533A", - "CRC32": "228DBE92", - "BLAKE3": "8AC0B12A8C87EC51BDD5403BD868881E31E895155143ED80E2621F2FC482552E", - "AutoV3": "22D61E2DAB89" - }, - "primary": true, - "downloadUrl": "https://civitai.com/api/download/models/2602723" - } - ], - "images": [ - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e5021552-1f38-4869-9ef8-961b31769671/original=true/117669784.jpeg", - "nsfwLevel": 8, - "width": 2742, - "height": 4096, - "hash": "U4Gu%jKk05~B00rC~qIo00?wxuRkjCRPHqNF", - "type": "image", - "metadata": { - "hash": "U4Gu%jKk05~B00rC~qIo00?wxuRkjCRPHqNF", - "size": 14776728, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "19岁韩国Kpop偶像coser，皮肤光滑细腻，通体雪白，极致冷白皮，脸蛋精致甜美如少女，象牙白冷白皮肤泛着亲密潮红，妆容干净可爱，棕色大地眼影柔和晕染，眼线细腻贴合眼形，睫毛浓密卷翘自然上扬，唇色玫瑰豆沙湿润带晶莹光泽，轻微咬唇痕迹显粉嫩齿痕，腮红饱和桃粉晕染成小猫脸庞。长直黑发顺滑披散至腰际，几缕发丝因汗水微微粘在脸颊和颈侧。脸型精致瓜子脸，妆容清透自然：大地色眼影轻晕染，眼线细致贴合双眼皮，浓密自然睫毛，唇色玫瑰豆沙带湿润光泽。身材丰满火辣，巨乳挺拔，腰肢纤细，臀部浑圆翘挺，全身赤裸，乳头粉嫩挺立，阴部光洁无毛。\n\n她微微嘟起粉嫩双唇，做出可爱嘘声姿势，食指竖在唇前，指尖轻轻抵住下唇中央，眼睛半眯成月牙形，眼尾上扬带点俏皮挑逗，瞳孔清亮反光明显，睫毛纤长自然卷翘，眉毛细而平直。右臂自然垂落贴着身体做支撑，左手抬起比出嘘的手势，上半身微微前倾，胸部自然下压形成明显弧度，腰部极度收紧，臀部向后翘起形成流畅曲线，脚尖朝向镜头，赤脚跪在浅灰色毛绒地毯上，脚趾微微蜷曲。\n\n室内现代简约卧室，中景清晰拍摄，人物占据画面70%面积，从胸口以上到大腿中部完整入镜，略低角度仰拍，背景是一个正在熟睡的女人，她大约30岁左右，皮肤白皙，长发散乱在枕头上，全身赤裸，床上铺着白色亚麻床单，床头柜上放着一盏简约台灯和一本半开的书，背景墙面浅灰白色挂着简易抽象画，右后方有模糊的白色窗帘和浅木色床头板，空间感由近处地毯绒毛→人物身体→中远景床上熟睡女人依次递进，形成自然的视觉引导路径。\n\n小红书纯欲风，自然窗光从画面左侧45度高位打来，主光柔和偏冷白，右侧有微弱环境补光形成淡淡轮廓光，人物鼻梁与锁骨高光明显，腹部与大腿内侧产生细腻阴影过渡，整体画面真实、通透，像用佳能R系列+50mm f/1.8在自然光下抓拍的纪实感人像，快门瞬间凝固的鲜活与性感并存。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2eb50140-e2d6-44c1-b1b8-edd1b3e6cb0a/original=true/118225768.jpeg", - "nsfwLevel": 8, - "width": 2750, - "height": 4888, - "hash": "U8ByaJ,@9GZ~~Xo3emoe4oadxaog9Gs.baRQ", - "type": "image", - "metadata": { - "hash": "U8ByaJ,@9GZ~~Xo3emoe4oadxaog9Gs.baRQ", - "size": 18402175, - "width": 2750, - "height": 4888 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "人物设定 & 发型设计 & OOTD 全解\n三位女性挤在一个狭小的电话亭内，均为19岁韩国童颜巨乳风格，皮肤雪白，细腻的肌肤纹理清晰。电话亭左边女性胸部最大，下垂严重，她留着紫色渐变长发，扎成高马尾。她穿着一件被雨水淋湿的纱质短袖T恤，T恤面料薄透，湿透后紧紧贴在皮肤上，完全透明，清晰地勾勒出她巨大下垂乳房的轮廓，T恤下摆卷起在胸下打了个结，露出整个腰腹，下身是湿透的浅蓝色牛仔短裤，裤裆部位紧贴阴部，她赤脚，但旁边放着一双透明矮跟凉鞋。中间女性胸部丰满下垂，留着黑色波波头短发。\n\n她只穿了一件 oversized 的灰色男士衬衫，衬衫材质为牛津纺，衣襟完全敞开，暴露出整个前身，包括她下垂的乳房、平坦的小腹和稀疏的阴毛，衬衫袖子挽起，下摆长及大腿中部，她未穿内衣裤。\n\n右边女性身材相对娇小但胸部仍显丰满，留着金色羊毛卷长发。她穿着一件白色紧身露脐针织衫，针织衫领口被故意拉至肩膀以下，露出一侧的肩膀和整个手臂，胸部部分紧绷，乳沟深邃，乳房下缘因下垂在布料下形成明显的弧线，下身是一条黑色皮质包臀迷你裙，裙子极短，紧紧包裹臀部，裙摆边缘陷入臀缝，她脚穿透明尖头高跟鞋。\n\n表情管理\n电话亭内空间拥挤，三人表情亲密而暧昧。左边女性脸颊紧贴中间女性的肩膀，眼神迷离地看着亭外，嘴唇微张呵出热气。中间女性低头看着右边女性的胸部，脸上带着顽皮的笑意，舌头舔着嘴角。右边女性抬头看着中间女性的脸，表情羞涩中带着大胆，眼神湿润。三人妆容略有晕染，更添真实感。\n\n肢体状态与细节\n在狭窄的电话亭内，三人身体紧贴。左边女性的湿透身体紧贴着中间女性的侧身，她的一只手扶在中间女性的腰侧，另一只手搭在电话机上。中间女性的手臂环抱着右边女性的肩膀，另一只手则看似无意地轻抚着右边女性从针织衫下拉裸露的腰肢。右边女性身体微微蜷缩，一只手挡在自己胸前，但更像是欲拒还迎，另一只手扶着电话亭的玻璃墙。她们的下半身紧紧相靠，腿部互相摩擦。\n\n场景与构图\n场景是一个复古的红色公用电话亭内部，外部是朦胧的雨夜红灯区街道。前景是电话亭的玻璃门，略有水珠和雾气，透过玻璃，三位女性是画面的绝对中心，清晰可见。中景是电话亭内部狭小的空间、金属电话机、她们纠缠的身体和散落的衣物鞋子。背景是电话亭外模糊的霓虹灯光、雨丝和流动的光斑。构图采用从电话亭外向内拍摄的视角，玻璃框形成画框，聚焦于内部香艳的场景。\n\n光影设计 &拍摄要点\n光源主要来自电话亭顶部的内置照明灯（冷白色）和亭外街道的彩色霓虹灯光（红色、蓝色）。内部顶光在女性们的头顶、肩膀和乳房上部形成高光，在眼窝、乳沟和下颚下方投下阴影。外部霓虹光透过淋湿的玻璃，在她们的身体侧面上映出斑斓的色块。湿透的衣物和皮肤产生复杂反光。拍摄模拟电影感镜头，使用大光圈（f/2.0）使亭外背景完全虚化成彩色光斑，焦点在人物面部和身体上，快门速度较慢，略微捕捉到雨丝轨迹，整体色调偏冷，但人物皮肤受内部灯光影响保持白皙，高感光度带来轻微颗粒感，增强氛围。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/cf4e6525-20c4-445d-97ea-5e2185856d17/original=true/118230710.jpeg", - "nsfwLevel": 4, - "width": 2304, - "height": 4096, - "hash": "U8F#:z9@00I:}]EKI.E101xs9t9F#Yxt5Q%3", - "type": "image", - "metadata": { - "hash": "U8F#:z9@00I:}]EKI.E101xs9t9F#Yxt5Q%3", - "size": 10760919, - "width": 2304, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "人物设定 & 发型设计 & OOTD 全解\n\n画面中央三位韩国女性，年龄均为十九岁，呈现典型的韩式网红审美风格。最左侧女性身高约一米六五，体型纤细，拥有极致的冷白肤色，肤质如瓷器般光滑，在光照下呈现出轻微过曝的亮白效果，皮肤表面细腻的毛孔与柔软的绒毛清晰可见。她长着一张童颜面孔，脸颊饱满，但胸部尺寸惊人，是丰满的G罩杯，乳房因重量而明显下垂，乳头的形状在衣物下清晰凸显，乳头与乳晕呈现淡粉色，与雪白肌肤形成对比。她留着及腰的波浪长卷发，发色为浅亚麻金。她身穿一件黑色漆皮材质的抹胸式超短连衣裙，材质硬挺光亮，抹胸部分采用平直线条裁剪，紧紧勒裹住胸部下缘，将双乳向上托起并挤压出深邃的乳沟，裙摆长度仅能勉强遮住臀部，下摆为不规则撕裂边设计。裙子侧边有细长的银色拉链。脚穿一双透明PVC材质的高跟鞋，鞋跟为十厘米细跟，鞋面完全透明，露出涂着鲜红色指甲油的脚趾。中间女性身高最矮，约一米五八，同样肤色雪白过曝，童颜稚嫩，但胸部为H罩杯，乳房更为硕大饱满，下垂程度更甚，乳房的自然垂坠形态明显。她梳着齐刘海黑长直发型，发丝笔直垂顺。她身穿一套玫红色亮片材质的比基尼式上衣与极短裤，上衣仅为两条细带支撑的三角杯，勉强覆盖住乳头，乳房大部分侧肉与下缘完全暴露在外，亮片由细密线迹缝制在弹性网布底衬上。短裤是低腰设计，裤腰边缘低于髋骨，裤腿极短，紧裹住臀瓣，臀沟隐约可见。脚上是同款透明PVC高跟鞋。右侧女性身高最高，约一米七，四肢极其纤细修长，肤色冷白，胸部为F罩杯，乳房呈水滴形下垂。她梳着高马尾发型，发色为深棕色。她身穿一件白色镂空钩花蕾丝连体衣，蕾丝为花卉图案镂空，透过网眼可直接窥见其下的雪白肌肤与粉色乳头，连体衣为深V领设计，开到胸骨下方，腰部完全镂空，仅靠侧边细带连接，下身是类似丁字裤的裁剪，仅以细窄的蕾丝条遮挡耻骨区域，臀部完全裸露。同样穿着透明高跟鞋。\n\n表情管理\n\n三位女性并排站立，身体略朝向镜头方向，但头部均微微低垂，目光聚焦于各自手中握持的智能手机屏幕。她们的面部妆容极度精致，底妆白皙无瑕，眉毛为平直眉形，眼妆采用粉棕色系眼影，搭配浓密卷翘假睫毛与精细眼线，佩戴浅灰色美瞳。嘴唇涂抹带有晶莹光泽的浅粉色唇釉，唇形饱满。表情呈现出沉浸于屏幕内容的淡漠与疏离，眼神低垂专注，嘴角无显著弧度，面部肌肉处于自然放松状态。\n\n肢体状态与细节\n\n她们站在人行道边缘，背对着后方店铺。左侧女性身体重心落在右脚，左脚尖轻点地面，膝盖微曲，左手持手机，右手食指在屏幕上滑动，右臂抬起导致腋下区域微微收紧，胸部因姿势更显集中。中间女性站姿较直，双腿并拢微微内八，双手捧握手机置于胸前下方，此姿势使其下垂的双乳向前聚拢，乳沟深邃，下缘弧线完全暴露。右侧女性身体略向右侧转，左腿伸直，右膝弯曲脚尖点地，臀部自然侧翘，右手操作手机，左手手指勾着带金属链条的黑色小手包垂于腿侧。她们裸露的冷白肌肤在光线下泛着光泽，乳房的自然下垂形态与纤细腰肢、四肢形成夸张对比，高跟鞋凸显小腿线条，脚背绷直。\n\n场景与构图\n\n镜头从马路中央略微仰视的角度拍摄，指向站在人行道上的三位女性。前景下方是马路的深灰色沥青路面，湿润反光，可能有一两道模糊的白色交通标线穿过。中景即三位女性，她们是画面绝对主体，清晰锐利，占据大部分画面，背对后方店铺。背景是霓虹闪烁的红灯区店铺街景：紧贴人物后方是一家悬挂着巨大桃红色心形霓虹灯招牌的酒吧，招牌周边缠绕着蓝色LED灯带，橱窗玻璃映出内部暖昧的红色灯光与晃动人影轮廓；右侧相邻是一家紫色霓虹灯管拼出“按摩”韩文字样的店面，门口垂挂珠帘，内部透出橙色光晕；左侧可见一家闪烁着白色与绿色霓虹的便利店招牌；更远处街道两侧排布着各色发光招牌，形成五彩斑斓的光带，并有模糊的行人身影与车辆灯光轨迹。空间关系明确，由近及远为：马路前景 → 人行道上的人物中景 → 紧贴人物的店铺立面背景 → 纵深街道远景，视觉引导路径从下方路面延伸至人物腿部，向上经过身体曲线，最终聚焦于面部与手机屏幕，背景的霓虹光芒作为氛围衬托。\n\n光影设计 &拍摄要点\n\n主要光源来自背景店铺密集的彩色霓虹招牌（桃红、蓝、紫、绿、橙色），以及街道上路灯的白色冷光。光线方向为从人物背后及侧后方射来，形成背光与侧逆光效果，强度中等偏强，色彩浓郁。实际影响：强烈的彩色背光在三位女性身体轮廓（尤其是头发、肩颈、手臂、臀部及腿部曲线）边缘勾勒出明亮的彩色光晕，创造出强烈的分离感；她们身体正面（面对镜头一侧）主要受环境漫反射光与手机屏幕冷光补光，亮度较低但足以展现细节，冷白肌肤与深色或艳色衣物在对比中更显突出；透明高跟鞋反射着后方霓虹光彩；马路路面反射着上方灯光形成光斑。拍摄模拟专业镜头大光圈夜景人像效果，使用f/1.4超大光圈获得极浅景深，使背景霓虹招牌与街道光轨虚化成绚丽多彩的圆形与条形光斑，人物主体保持极高锐度；感光度设为ISO 2500以捕捉暗部细节并维持快门速度，画面带有轻微噪点以增强真实感；白平衡稍偏冷，强化肤色冷白感与霓虹灯暖色的对比，整体呈现戏剧化又具现实感的夜间街拍质感。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/fbde5ca1-ea10-4649-9c01-826865608a3f/original=true/117669537.jpeg", - "nsfwLevel": 16, - "width": 2742, - "height": 4096, - "hash": "U8HBMYqZ%gQl?^rXtlIUR:4._3S$-pRj%No}", - "type": "image", - "metadata": { - "hash": "U8HBMYqZ%gQl?^rXtlIUR:4._3S$-pRj%No}", - "size": 15391958, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "人物设定 & 发型设计 & OOTD 全解\n19岁韩国Kpop偶像coser，皮肤光滑细腻，通体雪白，极致冷白皮，脸蛋精致甜美如少女，象牙白冷白皮肤光滑透亮泛着亲密潮红细腻光泽，妆容干净可爱，棕色大地眼影柔和晕染，眼线细腻贴合眼形，睫毛浓密卷翘自然上扬，唇色玫瑰豆沙湿润带晶莹光泽，轻微咬唇痕迹显粉嫩齿痕，腮红饱和桃粉晕染成小猫脸庞。长直黑发顺滑披散至腰际，几缕发丝因汗水微微粘在脸颊和颈侧。脸型精致瓜子脸，妆容清透自然：大地色眼影轻晕染，眼线细致贴合双眼皮，浓密自然睫毛，唇色玫瑰豆沙带湿润光泽。身材丰满火辣，巨乳挺拔，腰肢纤细，臀部浑圆翘挺，全身赤裸，乳头粉嫩挺立，阴部光洁无毛。男性为身材健硕的65岁白人男性老头，肌肉线条分明，腹肌清晰，阴茎粗壮勃起，青筋凸显，皮肤略带古铜色。\n\n表情管理\n女性嘴唇微张呈惊讶与渴望混合的O形，眼神向上仰视男性，瞳孔放大，带着一丝紧张与兴奋的俏皮。男性低头俯视女性，表情专注而充满欲望，嘴唇紧抿。\n\n肢体状态与细节\n女性跪坐在粉色床单上面对男性，双膝分开，臀部坐在脚跟上，上身前倾，双手握住男性勃起的阴茎，掌心包裹茎身，指尖轻轻扣住根部，指甲涂抹裸粉色。男性站立面对女性，双腿微分，左手搭在女性后颈，右手自然垂落。她的巨乳因前倾而垂坠晃动，乳晕清晰可见。\n\n场景与构图\n卧室内部，粉色床单凌乱铺展，白色窗帘背景模糊。中景为两人面对面的亲密互动，前景是女性握住阴茎的手部和垂落的长发丝，背景是柔和白色墙面与床头柜。镜头采用低角度仰视构图，强调女性仰头仰视的视角与男性俯视的压迫感，视觉引导从女性脸部向上延伸至男性阴茎。\n\n光影设计 & 拍摄要点\n柔和暖白自然光从右侧窗帘方向洒入，形成侧逆光，在女性脸颊、肩线、乳房轮廓勾勒出金边高光，男性身体左侧阴影加深肌肉立体感。床头灯补充微弱暖光，从上方偏左打下，在床单上形成浅浅阴影。画面呈现真实近距离手机拍摄质感，轻微颗粒感与自然色差，超高细节，8k，真实摄影风格。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0e8b27ef-6500-4196-8d96-4c4554299c4c/original=true/117668912.jpeg", - "nsfwLevel": 32, - "width": 2740, - "height": 4096, - "hash": "UMHUR#s+9@%MI]RiV@W=0foJxakD~VbFWBt7", - "type": "image", - "metadata": { - "hash": "UMHUR#s+9@%MI]RiV@W=0foJxakD~VbFWBt7", - "size": 15813121, - "width": 2740, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "主体是一位高挑的18岁东方美少女 反差婊 \n模特： 18岁少女，极致幼态脸型\n\n甜美的日系偶像脸，笑容极具感染力，身材娇小但比例极佳，腿型修长。发型设计： 粉色长发扎成高双马尾，佩戴着带有猫耳装饰的耳机（但未戴在头上）。OOTD全解： 一套 “电竞直播主”风格的粉色系制服印可爱pascale color字母\"MOODY\"。上衣是露脐短款卫衣，卫衣下摆被卷起卡在胸部上方，使得整个腹部和乳房裸露。卫衣内未穿内衣，乳房自然垂坠，乳尖朝向镜头。下身是超短百褶裙，裙下未着内裤，她正对着镜头（直播摄像头）做出“M字开腿”的经典动作，使得阴部完全暴露：阴毛修剪成心形，shaved pussy，阴蒂小巧可爱，大阴唇饱满敞开，阴道口完美展现，泛着水光。场景与拍摄构图设计： 大型动漫游戏展览（Comic-Con）的电竞舞台侧面休息区，身后是汹涌的参观人潮与炫目的广告屏。她坐在一张电竞椅上，面向自己的手机支架（镜头），身后是无数走动、驻足、投来目光的游客（动态模糊处理）。构图模拟她的自拍/直播视角，但暗示被身后路人用长焦偷拍。表情管理： 她对着“自己的镜头”摆出可爱的Wink和甜笑，但眼神的焦点似乎穿透了镜头，看向身后更广阔的人群，笑容中带着表演性的羞涩与实质性的炫耀。\n\n\n\n光影设计： 展馆内灯光杂乱，彩色LED屏光不断变幻写着“Moody V6\"。一道来自她侧后方（人群中偷拍者）的闪光灯，穿透嘈杂的环境光，精准地打在她分开的双腿之间，让私处的水光、心形阴毛的每一根都无比清晰，与她脸上表演性的可爱表情形成残酷对比。拍摄要点： 焦点清晰对准她的脸与暴露的阴部。背景人群处理成色彩斑斓的流动光斑，以突出主体。强调她表演性表情与真实身体状态的反差。捕捉“M字开腿”动作下，阴唇因拉扯而微微张开的形态。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/c65278da-0a9b-4507-ba0d-bc9d14f67a54/original=true/117669035.jpeg", - "nsfwLevel": 32, - "width": 2742, - "height": 4096, - "hash": "UAFrR^I.00-p?ER*R+M{03oM~BNHRSafn$V@", - "type": "image", - "metadata": { - "hash": "UAFrR^I.00-p?ER*R+M{03oM~BNHRSafn$V@", - "size": 16454052, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一位19岁的金发白人年轻女性，身材极端丰满且比例夸张到极致，胸部体积庞大沉重，乳房浑圆饱满向前隆起，几乎超出人体正常极限，皮肤呈现健康小麦色泽，肤质细腻平滑，带有真实细微毛孔纹理，完全哑光无任何油光或反光。她的头发是明亮亚麻金色长直发，发丝柔顺略带自然干草感，发尾轻微分叉，长度垂至乳房下缘，前额和脸侧散落几缕凌乱刘海，轻遮眉骨。妆容极淡，仅薄涂裸色唇膏，眼睫毛自然卷翘，眼白清亮，瞳孔深邃。\n她穿着一件正常设计的紧身浅芥末黄色短袖罗纹棉质T恤，面料薄而有弹性，细密均匀横向罗纹纹理清晰可见，标准圆领口与短袖双针车缝收边，本该完整包裹身体，此刻却被她双手从下摆位置猛力向上大幅拉扯，布料被极度拉伸变形，紧紧勒进乳房上缘，形成密集横向深褶与绷紧拉纹，T恤完全被扯到锁骨上方，彻底暴露出整对硕大雪白乳房，浅粉色乳晕完整清晰，乳头自然挺立突出，乳沟深邃立体，衣服本身没有任何暴露设计，全靠她主动拉扯造成瞬间闪现。搭配一条超短中腰水洗蓝色破洞牛仔热裤，裤腿仅盖臀部上沿，大腿根部大面积裸露，裤边毛边参差，多处手工撕裂破洞露出白皙内侧肌肤，裤腰纽扣解开一颗，拉链拉下一半，隐约露出白色蕾丝内裤细边。脚穿旧白色帆布鞋，鞋带松散。\n她站在体育用品商店狭窄货架通道正中央，正面直视镜头，头微微后仰，嘴唇微张，粉嫩舌尖轻触下唇，眼神混合挑逗与无辜的迷离。双手仍死死抓住T恤下摆向上猛提的动作完全定格，肘部大幅外展，锁骨深陷明显，腹部平坦带一丝自然软肉，圆润肚脐完全暴露。\n背景为杂乱仓储式货架，左侧密集排列绿色与橙色Prince网球拍及网球罐，右侧堆叠高耸棕色纸箱，箱面印粗黑英文“TRAINER”和“BOX1 of 1”，地面灰色水泥，远处天花板多排冷白荧光灯管。\n光线明亮均匀，主光源来自头顶正上方及偏右的多组强冷白荧光灯管，形成清晰明亮的顶光效果，脸部、乳房上半部及乳沟上缘获得强烈高光，皮肤呈现自然哑光白皙质感，无任何油腻闪光，乳房下半部与腹部保留柔和渐变阴影，整体画面明亮锐利，色调偏冷带轻微青白，景深较浅，背景货架轻度虚化，前景女性身体细节极度清晰锐利，带有Iphone 16 Pro Max 手机瞬间抓拍与超市监控的粗粝真实质感，定格在这一刹那的彻底暴露与禁忌挑衅张力。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/eceba931-bcde-4d30-8fff-60267e9b4ac5/original=true/117669131.jpeg", - "nsfwLevel": 8, - "width": 2742, - "height": 4096, - "hash": "U6Dlf:s%5n$yIWM_EM%100s8^ks:?Hxu~qD%", - "type": "image", - "metadata": { - "hash": "U6Dlf:s%5n$yIWM_EM%100s8^ks:?Hxu~qD%", - "size": 13512209, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一位年轻白皙的俄罗斯裔女性，约22岁，身材匀称丰满，皮肤细腻光滑带有轻微自然雀斑，胸部饱满自然下垂，乳晕浅粉色偏大，乳头挺立，腰肢柔软，小腹平坦带浅浅马甲线，肚脐镶嵌闪亮银色水滴形肚脐钉，阴部完全无毛，阴唇饱满外翻呈浅粉褐色，阴蒂微微外露，整体体毛极少仅腋下与小腿有极淡金色绒毛。长直深棕色头发中分，发丝柔顺垂至胸下乳沟位置，发尾自然内扣，额前碎发轻贴脸颊。完全赤裸，只佩戴细银色细链项链与左上臂细黑皮革手环，右前臂靠近肘部有小片棕榈树刺青。\n她坐在深绿色绒面台球桌上，双腿大幅度分开呈M字坐姿，膝盖弯曲，双脚自然垂落踩在球桌边缘，脚趾自然蜷曲，脚底朝向镜头。双手向后撑住桌面，手掌撑在两颗彩球之间，指尖轻触球面。脊背微微后仰，胸部因姿势自然前挺，乳房因重力轻微向两侧外扩。\n表情平静中带一丝慵懒挑逗，直视镜头，琥珀色眼睛半眯，眼尾上挑，睫毛浓密纤长，嘴唇自然微张，唇色玫瑰豆沙，嘴角带极淡不易察觉的弧度。\n画面采用中低机位略仰拍视角，台球桌占据前景至中景，深绿色绒布桌面清晰可见细密绒毛纹理与轻微磨损痕迹，散落着几颗彩色台球（橙色1号、绿色6号、紫色12号、红白相间8号黑球等）随意滚落在她大腿根部附近与脚边，形成视觉引导。背景是现代室内休闲空间，暗绿色沙发、浅绿抱枕与黑色墙面模糊虚化，墙上挂几盆绿植，整体空间纵深感强。\n光源主要来自左侧上方45度暖白色柔光灯，右侧有微弱环境冷蓝色补光，形成明显明暗对比，左侧脸颊与胸部左侧获得明亮高光，右侧身体陷入柔和阴影，乳沟与腹部形成立体阴影过渡，桌面绒布受光面呈现丝绒质感反光，球体表面有清晰高光点与环境反光。整体色调偏电影感冷暖对比，肤色真实自然，细节锐利，类似高端时尚艺术裸体摄影风格，使用35mm全画幅镜头f/2.8大光圈浅景深拍摄。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9d7eb0cc-ba28-41ec-848f-3fdb8570b97a/original=true/117669170.jpeg", - "nsfwLevel": 8, - "width": 2742, - "height": 4096, - "hash": "U9E2qfv{0Mt8^-$#?HS400-o%1S%nnaJ9ZkX", - "type": "image", - "metadata": { - "hash": "U9E2qfv{0Mt8^-$#?HS400-o%1S%nnaJ9ZkX", - "size": 15404493, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "lost it ;(" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/595c3b9b-d5e0-4404-aa81-f20eefc1f116/original=true/117669112.jpeg", - "nsfwLevel": 8, - "width": 2496, - "height": 3728, - "hash": "UYH1VoNGkWWB~VNG%1s.?bs.t6j??aayfks:", - "type": "image", - "metadata": { - "hash": "UYH1VoNGkWWB~VNG%1s.?bs.t6j??aayfks:", - "size": 13211952, - "width": 2496, - "height": 3728 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "一位二十五岁左右的亚洲女性，身材爆炸性丰满，胸围骇人（罩杯远超K杯），皮肤白皙如瓷泛蜜糖暖光，毛孔细腻近乎消失，锁骨深邃成沟，腰身纤细到极端沙漏比例，臀部饱满翘挺，大腿粗壮肉感，小腿意外修长。她一头浓密乌黑长直发如黑绸瀑布垂至臀下五厘米，发尾齐平微内扣，头顶扎一枚小型豹纹丝巾蝴蝶结，额前碎刘海被汗湿几缕贴额，脸侧发丝被风吹乱几根凌乱飞舞。\n她上身仅穿极致暴露的豹纹细带吊带抹胸，面料为超薄高弹丝绒质感莱卡混纺，触感柔软贴肤带细腻绒毛，经典豹纹印花黑底金棕色斑点，斑点大小不均边缘自然晕染。吊带极细仅一厘米宽，从颈后绕过再到胸前交叉，形成深V形领口，V底直达肚脐上方两厘米，布料被巨乳撑到极限出现密集放射褶皱与轻微拉丝纹理，乳沟深不见底，下半球乳房与乳头完全裸露挺立在外，乳晕粉嫩边缘在夕阳下泛着柔光。抹胸中央缝嵌一颗硕大水晶豹头吊坠，垂在乳沟正中随呼吸晃动。腰部以下完全赤裸，无任何布料遮挡，阴阜光滑无毛，阴唇饱满粉嫩因姿势轻微分开，阴蒂微微凸起，大腿根部皮肤紧致泛蜜色光泽，臀部与臀缝全部裸露，腰侧各垂两条长二十五厘米金色细链条作为唯一装饰，链尾坠小豹爪吊坠，轻晃时反射金光。\n她神情放肆魅惑，嘴角坏坏上扬露出雪白贝齿与粉舌轻吐，眼睛眯成狡黠月牙，眼尾上挑画浓黑眼线，浓密卷翘假睫毛投下扇形阴影，瞳孔琥珀色泛金光，带着肆无忌惮的挑逗与征服欲。\n她坐在非洲草原干燥黄褐色草地上，双腿大幅打开呈M形，右腿自然伸直脚尖点地，左腿大幅抬起弯曲，膝盖向胸前拉近，小腿垂直向上，脚掌朝天，脚趾绷直，脚踝线条优美，大腿内侧肌肉因抬腿姿势绷紧显出性感线条。右手随意撑在身后草丛中，左手轻轻搭在一只成年非洲豹的肩背上，手指轻抚豹子颈侧斑纹毛发，豹子慵懒侧卧在她左侧紧贴大腿，头部微微抬起，琥珀色豹眼半眯直视镜头，鼻尖几乎碰到她大腿内侧，呼吸间热气若隐若现。另一只豹子从右后方靠近，鼻尖贴近她臀侧，第三只豹子在前方两米处蹲坐，目光锁定她全身，形成极度亲密包围感。私处因双腿大开与抬腿姿势彻底暴露，细节一览无遗。\n画面为中近景构图，她占据画面80%面积，身体正对镜头充满压迫感，草地与灌木作为中景占据画面下三分之一，三只健硕成年非洲豹环绕她，形成亲密互动包围，最近的两只几乎贴身，第三只稍远，背景是开阔草原与远方金色夕阳余晖下的稀疏树影。空间由她的身体与豹子强烈拉向前景→草丛中景→远方草原层层推进，视觉冲击与亲密感极强，完美融合野外豹群环境。\n黄金时段炽热橙金夕阳从画面右后方50度角斜射，强逆光在她发丝、身体轮廓与豹子毛边勾出耀眼金边，脸部与胸部右侧爆闪高光，左侧落入性感深影，豹纹绒面与真豹毛被光线打出立体绒毛质感，整个画面色温极暖，真实非洲草原野生纪实摄影感，带轻微手机镜头眩光、尘埃颗粒与细颗粒噪点，像专业模特在野外与豹群亲密互动时用高端手机抓拍的惊艳瞬间。" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - }, - { - "url": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/89676511-f0cf-4e02-9cda-20e8271bf2ba/original=true/117669996.jpeg", - "nsfwLevel": 8, - "width": 2742, - "height": 4096, - "hash": "U8D+6+%ME2~BHr4oIUnh0K~W~Ciw0LIAMdoz", - "type": "image", - "metadata": { - "hash": "U8D+6+%ME2~BHr4oIUnh0K~W~Ciw0LIAMdoz", - "size": 16503174, - "width": 2742, - "height": 4096 - }, - "minor": false, - "poi": false, - "meta": { - "prompt": "第一人称约会视角，面对面坐在昏暗高级餐厅角落卡座沙发，女孩坐在对面，身体微微前倾伸手拿菜单时，超短裙自然上滑，腿部不经意微张。25岁黑人女孩，深巧克力色皮肤细腻油润带性感珠光，妆容精致热辣，小猫眼眼尾上挑敷暖金棕闪片眼影，睫毛浓密卷翘，眼眸深邃带水雾，饱满厚唇涂玫瑰红水光唇釉轻轻抿着，腮红暖桃粉晕染成自然潮红。黑色紧致4C卷发中分，刘海微卷散落额头，几缕卷发贴在汗湿锁骨与丰满胸前，夸张银色环状耳环。身材极度丰满火辣，超大胸围将浅米白色低胸缎面紧身上衣撑到极限，领口深V露出深邃乳沟与胸部上半饱满弧线，布料紧绷勒出乳房侧边曲线，下身穿黑色皮质超短A字裙，裙长仅盖住臀部一半，她前倾时裙摆被沙发边缘带起向上卷起，双腿微张一条缝，露出无内裤私处隐秘一角，深色饱满外阴唇自然闭合，粉褐内阴唇边缘若隐若现，阴蒂小珠隐约凸起，像坐姿不稳的意外。臀部极圆翘大，裙子被撑得紧绷，侧面可见丰满臀肉溢出裙边。双手轻握菜单，指尖涂裸粉色，指节微微泛白。桌面有两杯红酒与蜡烛，前景是她前倾胸部挤压乳沟与裙底隐约走光私处边缘，背景是暖红壁灯与模糊帘子。烛光从桌面向上打，在乳沟与私处形成柔和高光与深邃阴影，眼眸抬头带一丝无察觉的羞涩，真实约会不经意含蓄走光瞬间，小红书丰满性感风，高度写实，细节饱满，8k" - }, - "availability": "Public", - "hasMeta": true, - "hasPositivePrompt": true, - "onSite": false, - "remixOfId": null - } - ], - "downloadUrl": "https://civitai.com/api/download/models/2602723", - "creator": { - "username": "catlover1937", - "image": "https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/18485719-777f-43e9-b4ae-7ae3000e5468/width=96/catlover1937.jpeg" - } - }, - "tags": [ - "nsfw", - "person", - "sexy", - "nude", - "girls", - "porn", - "asian", - "woman", - "realistic", - "photorealisic", - "base model", - "sex" - ], - "modelDescription": "I just made a new Telegram Group, drop in for chats, questions, feedbacks or just to share your work. I will be sharing prompts and pics there as CIVITAI has been removing my work repeatedly.
2026-01-20
New simple workflow available in case you need one.
2026-01-16
Hi — V6 is finally here. I’ve been pushing hard to get this out as soon as possible, but it just didn’t feel ready. After a full week of endless testing, tweaking, and merging, I finally feel confident saying this is the best version so far — and a huge improvement over V5.
What’s new in V6:
Updated to use MysticXXX V5 in conjunction with V4, since V5 alone is quite destructive to the existing model
Removed several redundant LoRAs, or ones that only introduced disadvantages to the model
Rebased the model on the zImage Turbo base, making LoRA usage much easier and producing images that stay closer to the base model compared to V5
Huge improvement on genitalias
This version represents the most refined and stable state of the model to date. Hope you enjoy it!
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2026-01-09
After few sleepless nights and nonstop fine-tuning and testing, V5 IS HERE, extensive refinement and fine-detail adjustments were completed,
Found a working way to replace Mystic XXX V3 with V4.
Removes the unpleasant yellow color cast that affected outputs in V3.
Fixes pose anomalies when characters are farther from the camera.
Significantly improves XXX poses.
Further enhances compatibility with character LoRAs.
PS. Guys stop frigging' reporting my images, all my prompts start with 19 years old woman, and NOT friggin' minors. You know what, from now on you will only see me use \"48 years old mature lady\" instead :(
2026-01-07
Use V3 for consistency, V4 if you want to try something new. I am actively working on V5.
2026-01-06
HAPPY NEW YEAR! ZIT V4!
A lot of fine-tuning went into this version with the goal of improving compatibility with other LoRAs (especially character LoRAs), while preserving what V3 already does well. That said, please remember this is already an all-in-one checkpoint, so keep your LoRA weights very light. In my tests shown in the carousel, I used around 0.65 for character LoRAs and it worked quite well.
I experimented with MysticXXX V4, but it ended up breaking compatibility with other LoRAs. Once MysticXXX V4 is applied, adding additional LoRAs tends to cause distortions, so I decided to keep MysticXXX at V3 instead.
Baked in LSP to unlock more XXX poses 👏 big thanks to the creator, an absolute legend. However, One side effect is that it can drift quite a bit from the original faces. I’ve spent time trying to find a good balance, and I think the trade-off is worth it, but I’d love to hear what you think.
2025-12-25
Merry XMAS! ZIT V3 is here now
MysticXXX has updated their model to V3, so this release is now based on that version.
I’ve optimized the LoRA weights according to my preferences, with the hope that it’s more lightweight and compatible with additional LoRAs.
The FP8 version is also tested and up running OK.
PS: I’ve seen a few comments mentioning image breakage when stacking multiple LoRAs. I’ve noticed this in my own generations as well. That said, this checkpoint is an all-in-one, ready-to-use model, so you generally shouldn’t need to add extra LoRAs on top. Hopefully, with the V3 update and optimizations, images will be less distorted when LoRAs are used—but we’ll see.
Anyway, good luck generating! Feel free to share your results below—I’d love to see how others are using this model.
2025-12-24
Hi again after a bit more than a year,
I am annoyed of adding tones of lora every time when I generate so i merge this checkpoint.
I mixed RedCraft | 红潮造相 with turbo base as my base. Then merge DetailDeamonZ and [ZIT] Mystic XX PLUS 2 of the loras I trained, one gives better asian faces (to my own taste of course <3), nudes and nipples, one gives lighter skin (but not the pale and textureless one the turbo base has).
Why isn't there V1?
Because I never thought of sharing until someone asked, and I lost V1 :sadge:
Use at your own risk :)
PS. I am too lazy to add prompts, if you want them send me a DM.
SD 1.5
I was always looking for a model that can draw good nsfw photos with detailed private parts, so I made this for fun. Hope you like it.
This checkpoint generates very NSFW content. Put NSFW in the negative prompt if needed.
v4-beta: did some merging with my other model, and updated face model and smoothen the skin texture. Some people might like it while some don't, stick with v3 if you don't like the update.
2024-09-18: Re-uploaded the V3 pruned, seems to be working now. Sorry for the mess :(.
v3: Greatly improve model stability, better compactivities with NSFW Loras. Might look quite different from v2
Full/Pruned version available for v3
v2: Improved skin tone and quality of hands and faces as comment suggested.
Full/Pruned version available for v2
v1: initial release
Guidance: 7-15
Steps: 25 - 35
Sampler: DPM++ 2M
Hi-res fix 2x recommended. Adetailer on face was used.
", - "civitai_deleted": false, - "favorite": false, - "exclude": false, - "db_checked": false, - "metadata_source": "civitai_api", - "last_checked_at": 1769145853.084197, - "usage_tips": "{}", - "model_type": "checkpoint" -} \ No newline at end of file diff --git a/zimageturbo/moodyPornMix_zitV6.png b/zimageturbo/moodyPornMix_zitV6.png deleted file mode 100644 index 7a1e5dee47d5cfc9d45659cf3e4d130fc5fb276e..0000000000000000000000000000000000000000 --- a/zimageturbo/moodyPornMix_zitV6.png +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7cc2898516feb41560318719b0e1d9556c22b6d42e61809a6ed35eea0dd1529c -size 14776728