Upload model using ModelScope SDK

.gitattributes

@@ -44,10 +44,10 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.tar filter=lfs diff=lfs merge=lfs -text
 *.wasm filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -textassets/MOSS-Image-Base/dit_4.5bit.jpg filter=lfs diff=lfs merge=lfs -text
-assets/MOSS-Image-Base/dit_int8_6bit.jpg filter=lfs diff=lfs merge=lfs -text
-assets/MOSS-Image-SFT/bf16.jpg filter=lfs diff=lfs merge=lfs -text
-assets/MOSS-Image-SFT/dit_3.5bit.jpg filter=lfs diff=lfs merge=lfs -text
-assets/MOSS-Image-SFT/dit_4.5bit.jpg filter=lfs diff=lfs merge=lfs -text
-assets/MOSS-Image-SFT/dit_6bit.jpg filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+"/assets/MOSS-Image-Base/dit_int8_6bit.jpg" filter=lfs diff=lfs merge=lfs -text
+"/assets/MOSS-Image-Base/dit_int8_7bit.jpg" filter=lfs diff=lfs merge=lfs -text
+"/assets/MOSS-Image-SFT/dit_int8_6bit.jpg" filter=lfs diff=lfs merge=lfs -text
+"/assets/MOSS-Image-SFT/dit_int8_7bit.jpg" filter=lfs diff=lfs merge=lfs -textassets/MOSS-Image-Base/dit_int8_6bit.jpg filter=lfs diff=lfs merge=lfs -text
 assets/MOSS-Image-SFT/dit_int8_6bit.jpg filter=lfs diff=lfs merge=lfs -text
+assets/MOSS-Image-SFT/dit_int8_7bit.jpg filter=lfs diff=lfs merge=lfs -text

README.md

@@ -4,41 +4,49 @@ We provide pre-quantized weight packages produced by [**EntroPack**](https://git
 
 The packages run with as little as 4 GB of VRAM.
 
-Base is provided at three levels, SFT additionally ships a 3.5-bit variant, and the 3.5-bit and 4.5-bit DiTs use mixed quantization. Base and SFT share the same text encoder packages:
+Base is provided at four levels, SFT additionally ships a 3.5-bit variant. Base and SFT share the same text encoder packages.
 
-* **`3.5bit`** (SFT only, lowest VRAM) — mixed-bit: the quantization-sensitive layers keep **5.0 bpp**, while the remaining layers use **3.2516 bpp** (weighted target 3.5 bpp);
-* **`4.5bit`** (balanced) — mixed-bit: the quantization-sensitive layers keep **5.0 bpp**, while the remaining layers use **4.4172 bpp** (weighted target 4.5 bpp);
-* **`6bit`** (highest precision) — uniform entropy coding at 6 bits per weight;
-* **`int8_6bit`** (fastest inference) — weights are first **INT8 W8A8-quantized**, then entropy-coded by EntroPack down to 6-bit storage; the forward pass uses **INT8 GEMM**, which delivers a faster inference speed than the BF16 GEMM used by the other levels.
+**BF16 GEMM levels** (the 3.5-bit and 4.5-bit DiTs use mixed quantization):
+
+* **`3.5bit`** (SFT only, lowest VRAM) — mixed-bit: the quantization-sensitive layers keep **5.0 bpp**, while the remaining layers use **3.2516 bpp**, giving **3.526 bpp** on the package;
+* **`4.5bit`** (balanced) — mixed-bit: the quantization-sensitive layers keep **5.0 bpp**, while the remaining layers use **4.4172 bpp**, giving **4.545 bpp** on the package;
+* **`6bit`** (highest precision of the BF16 GEMM levels) — uniform entropy coding, **6.055 bpp** on the package.
+
+**INT8 GEMM levels** use a hybrid allocation:
+
+* **`int8_6bit`** (fastest inference) — the quantization-sensitive linears including embeddings, time embeddings, modulation and final layers are entropy-coded at **8.5 bits per weight** and run on **BF16 GEMM**, while all remaining linears are **INT8 W8A8-quantized with a ConvRot rotation** and run on **INT8 GEMM**; the package averages **5.999 bpp**;
+* **`int8_7bit`** (highest precision of the INT8 levels) — the same hybrid allocation, averaging **7.004 bpp** on the package.
+
+The INT8 GEMM path requires Triton and a CUDA GPU with compute capability 8.0 or newer.
 
 All packages are entropy-coded from the original BF16 weights directly. Every `*.safetensors` ships with a same-named `*.json` that records its quantization config. `actual bpp = 8 × Σ storage bytes / Σ numel` is measured on the package artifacts.
 
 ## Preview
 
-All previews are generated with the same prompt and seed; `bf16` is the full-precision reference.
+All previews are generated with the same prompt and seed; `bf16` is the full-precision reference. The quantized previews are ordered by PSNR against the `bf16` reference (highest first).
 
 ### SFT
 
 <table width="100%">
   <tr>
     <td width="33%" align="center">bf16 (reference)</td>
-    <td width="33%" align="center">6bit</td>
+    <td width="33%" align="center">int8@7</td>
     <td width="33%" align="center">int8@6</td>
   </tr>
   <tr>
     <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/bf16.jpg"></td>
-    <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_6bit.jpg"></td>
+    <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_int8_7bit.jpg"></td>
     <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_int8_6bit.jpg"></td>
   </tr>
   <tr>
+    <td width="33%" align="center">6bit</td>
     <td width="33%" align="center">4.5bit</td>
     <td width="33%" align="center">3.5bit</td>
-    <td></td>
   </tr>
   <tr>
+    <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_6bit.jpg"></td>
     <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_4.5bit.jpg"></td>
     <td align="center"><img width="100%" src="assets/MOSS-Image-SFT/dit_3.5bit.jpg"></td>
-    <td></td>
   </tr>
 </table>
 
@@ -46,16 +54,24 @@ All previews are generated with the same prompt and seed; `bf16` is the full-pre
 
 <table width="100%">
   <tr>
-    <td width="25%" align="center">bf16 (reference)</td>
-    <td width="25%" align="center">6bit</td>
-    <td width="25%" align="center">4.5bit</td>
-    <td width="25%" align="center">int8@6</td>
+    <td width="33%" align="center">bf16 (reference)</td>
+    <td width="33%" align="center">6bit</td>
+    <td width="33%" align="center">int8@7</td>
   </tr>
   <tr>
     <td align="center"><img width="100%" src="assets/MOSS-Image-Base/bf16.jpg"></td>
     <td align="center"><img width="100%" src="assets/MOSS-Image-Base/dit_6bit.jpg"></td>
-    <td align="center"><img width="100%" src="assets/MOSS-Image-Base/dit_4.5bit.jpg"></td>
+    <td align="center"><img width="100%" src="assets/MOSS-Image-Base/dit_int8_7bit.jpg"></td>
+  </tr>
+  <tr>
+    <td width="33%" align="center">int8@6</td>
+    <td width="33%" align="center">4.5bit</td>
+    <td></td>
+  </tr>
+  <tr>
     <td align="center"><img width="100%" src="assets/MOSS-Image-Base/dit_int8_6bit.jpg"></td>
+    <td align="center"><img width="100%" src="assets/MOSS-Image-Base/dit_4.5bit.jpg"></td>
+    <td></td>
   </tr>
 </table>
 
@@ -68,7 +84,8 @@ All previews are generated with the same prompt and seed; `bf16` is the full-pre
 | dit_3.5bit (mixed) | `models/MOSS-Image-SFT/dit_3.5bit.safetensors` | 3.526 | 2.24 GB |
 | dit_4.5bit (mixed) | `models/MOSS-Image-SFT/dit_4.5bit.safetensors` | 4.545 | 2.89 GB |
 | dit_6bit | `models/MOSS-Image-SFT/dit_6bit.safetensors` | 6.055 | 3.85 GB |
-| dit_int8_6bit (INT8 GEMM) | `models/MOSS-Image-SFT/dit_int8_6bit.safetensors` | 6.063 | 3.85 GB |
+| dit_int8_6bit (hybrid INT8 GEMM + ConvRot) | `models/MOSS-Image-SFT/dit_int8_6bit.safetensors` | 5.999 | 3.81 GB |
+| dit_int8_7bit (hybrid INT8 GEMM + ConvRot) | `models/MOSS-Image-SFT/dit_int8_7bit.safetensors` | 7.004 | 4.45 GB |
 
 ### Base
 
@@ -76,7 +93,8 @@ All previews are generated with the same prompt and seed; `bf16` is the full-pre
 |---|---|---|---|
 | dit_4.5bit (mixed) | `models/MOSS-Image-Base/dit_4.5bit.safetensors` | 4.545 | 2.89 GB |
 | dit_6bit | `models/MOSS-Image-Base/dit_6bit.safetensors` | 6.055 | 3.85 GB |
-| dit_int8_6bit (INT8 GEMM) | `models/MOSS-Image-Base/dit_int8_6bit.safetensors` | 6.063 | 3.85 GB |
+| dit_int8_6bit (hybrid INT8 GEMM + ConvRot) | `models/MOSS-Image-Base/dit_int8_6bit.safetensors` | 5.999 | 3.81 GB |
+| dit_int8_7bit (hybrid INT8 GEMM + ConvRot) | `models/MOSS-Image-Base/dit_int8_7bit.safetensors` | 7.004 | 4.45 GB |
 
 ## Text Encoder
 
@@ -86,7 +104,8 @@ The text encoder packages are shared by Base and SFT.
 |---|---|---|---|
 | text_encoder_4.5bit | `models/text_encoder_4.5bit.safetensors` | 4.552 | 5.97 GB |
 | text_encoder_6bit | `models/text_encoder_6bit.safetensors` | 6.070 | 7.29 GB |
-| text_encoder_int8_6bit (INT8 GEMM) | `models/text_encoder_int8_6bit.safetensors` | 6.075 | 7.29 GB |
+| text_encoder_int8_6bit (hybrid INT8 GEMM + ConvRot) | `models/text_encoder_int8_6bit.safetensors` | 6.000 | 7.23 GB |
+| text_encoder_int8_7bit (hybrid INT8 GEMM + ConvRot) | `models/text_encoder_int8_7bit.safetensors` | 6.993 | 8.08 GB |
 
 ## Usage
 
@@ -104,10 +123,10 @@ cd entropack && pip install -e '.[cuda13]'
 
 ### Example scripts
 
-* Inference: [standard](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_inference/MOSS-Image-EntroPack.py) / [low VRAM](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_inference_low_vram/MOSS-Image-EntroPack.py) — defaults to SFT 4.5bit; switch `dit_package` / `base_model_id` together to use Base or another level
+* Inference: [standard](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_inference/MOSS-Image-EntroPack.py) / [low VRAM](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_inference_low_vram/MOSS-Image-EntroPack.py) — defaults to SFT INT8 6bit; switch `dit_package` / `base_model_id` together to use Base or another level
 * LoRA training: [train](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_training/lora/MOSS-Image-entropack.sh) / [validation](https://github.com/modelscope/DiffSynth-Studio/blob/main/examples/moss_image/model_training/validate_lora/MOSS-Image-entropack.py) — defaults to SFT; replace `MOSS-Image-SFT` with `MOSS-Image-Base` in the dataset and model paths to train on Base
 
-`dit_package` and `text_encoder_package` are chosen independently: any combination works (e.g. `dit_int8_6bit` with `text_encoder_4.5bit`). The 6-bit packages are pre-registered in DiffSynth-Studio, so they load with a plain `ModelConfig`; the helper below registers any package's json sidecar before loading:
+`dit_package` and `text_encoder_package` are chosen independently: any combination works (e.g. `dit_int8_6bit` with `text_encoder_4.5bit`). The 6-bit packages are pre-registered in DiffSynth-Studio, so they load with a plain `ModelConfig`; the helper below registers any package's json sidecar before loading (the INT8 packages also read their ConvRot settings and 8.5-bit sensitive-layer allocation from it):
 
 ```python
 import json
@@ -147,9 +166,9 @@ pipe = MOSSImagePipeline.from_pretrained(
 )
 
 image = pipe(
-    prompt="a cat sitting on a windowsill",
+    prompt="""Full-color cinematic outdoor night photograph of a stunning young Asian woman in her early twenties with a natural warm ivory complexion and a healthy glow, wearing a soft blush pink off-shoulder dress with a fitted bodice, framed from the waist up against softly illuminated historic architecture. She stands near the centre of the frame, facing the camera with a gentle closed-lip smile, her hands resting loosely clasped in front of her waist with relaxed elbows; she has a slim oval face with softly defined cheekbones and a delicate narrow jaw, dark expressive eyes, natural-shaped brows, and subtle natural makeup with soft red lips. Her long black hair falls in loose natural waves over her shoulders, with fine strands framing her face. The fabric of the dress falls in soft folds across her shoulders and chest. In the frame-left background, the word "EntroPack" glows as a hand-drawn bubbly yellow neon sign mounted on a softly floodlit dark stone building above the street; rows of warm-lit windows, stone cornices and shopfronts remain clearly readable around the sign in gentle amber light, while the thick rounded irregular letters ride a gently bouncing baseline with a bright core and a soft outer glow, kept sharp and unblurred against the night. The background recedes into deep blue tones: the Eiffel Tower rises above distant buildings on the frame-right as a softly silhouetted tapering lattice outline in the shallow depth of field, while small amber, muted turquoise and rose-coloured lights dissolve into rounded bokeh across the upper background, leaving a darker area behind her head. Soft neutral frontal illumination with gentle falloff reveals her face with natural warm skin tone and subtle texture, a faint cooler rim separates her hair from the night, and the skirt of her dress fills the lower edge of the frame.""",
     height=1024, width=1024,
-    seed=42, num_inference_steps=50, cfg_scale=3.0,
+    seed=123, num_inference_steps=50, cfg_scale=3.0,
 )
 image.save("image.jpg")
 ```

models/MOSS-Image-Base/dit_int8_6bit.json

@@ -1,14 +1,66 @@
 {
-  "model_hash": "36922e1c521a9873c7d507512d25ab2e",
+  "model_hash": "955b29367fb6e506f5fb6b3acd50c1ed",
   "model_name": "moss_image_dit",
   "model_class": "diffsynth.models.moss_image_dit.MOSSImageDiT",
   "quant_config": {
-    "method": "entropack_lossy_quant_int8",
     "load_prequantized": true,
-    "backend_config_kwargs": {
-      "target_bpp": 6.0,
-      "row_rdo_iterations": 2,
-      "row_rdo_candidates": 20
-    }
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 5.782,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      }
+    ]
   }
 }

models/MOSS-Image-Base/dit_int8_6bit.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ebb384ba6527dd10ce7d7eeaaeb8a94718b58cd99d3a9f59f9e61d9ca0d5e13
-size 3852165592
+oid sha256:b94917cab4f43b56d5356b5bfbea747d1ff0910390f01aef1cca1434cdd211e9
+size 3811723984

models/MOSS-Image-Base/dit_int8_7bit.json

@@ -0,0 +1,66 @@
+{
+  "model_hash": "784655dc0c006e3ef6a3e89fa2975f7e",
+  "model_name": "moss_image_dit",
+  "model_class": "diffsynth.models.moss_image_dit.MOSSImageDiT",
+  "quant_config": {
+    "load_prequantized": true,
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 6.854,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      }
+    ]
+  }
+}

models/MOSS-Image-Base/dit_int8_7bit.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:225c783fcd6a0f0af10ad90771942c5788e7b4c07b6e985cab4e0637131d2e62
+size 4449823136

models/MOSS-Image-SFT/dit_int8_6bit.json

@@ -1,14 +1,66 @@
 {
-  "model_hash": "ad5f9335f74b2eb9d439ed118b0625ac",
+  "model_hash": "060367613bb0ebec4631f137437bfeb8",
   "model_name": "moss_image_dit",
   "model_class": "diffsynth.models.moss_image_dit.MOSSImageDiT",
   "quant_config": {
-    "method": "entropack_lossy_quant_int8",
     "load_prequantized": true,
-    "backend_config_kwargs": {
-      "target_bpp": 6.0,
-      "row_rdo_iterations": 2,
-      "row_rdo_candidates": 20
-    }
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 5.782,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      }
+    ]
   }
 }

models/MOSS-Image-SFT/dit_int8_6bit.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c16c9bde987116e76d042ba1b6a70fff633bdb5ba402aeaf360191e6184d7ca
-size 3852155752
+oid sha256:0ac2bb3ec2b2f02eb56e71aaabe91faf5cedfae138fd9403f16ac980e9672b44
+size 3811724464

models/MOSS-Image-SFT/dit_int8_7bit.json

@@ -0,0 +1,66 @@
+{
+  "model_hash": "2a95c50f3b178e8f49404604c1b8426e",
+  "model_name": "moss_image_dit",
+  "model_class": "diffsynth.models.moss_image_dit.MOSSImageDiT",
+  "quant_config": {
+    "load_prequantized": true,
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 6.854,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "video_embedder",
+          "text_embedder.proj",
+          "text_embedder.blocks.0.qkv",
+          "text_embedder.blocks.0.proj",
+          "text_embedder.blocks.0.ffn_gate",
+          "text_embedder.blocks.0.ffn_up",
+          "text_embedder.blocks.0.ffn_down",
+          "text_embedder.blocks.1.qkv",
+          "text_embedder.blocks.1.proj",
+          "text_embedder.blocks.1.ffn_gate",
+          "text_embedder.blocks.1.ffn_up",
+          "text_embedder.blocks.1.ffn_down",
+          "time_embedding.0",
+          "time_embedding.2",
+          "shared_block_modulation",
+          "final_linear_video",
+          "video_head_modulation"
+        ]
+      }
+    ]
+  }
+}

models/MOSS-Image-SFT/dit_int8_7bit.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a8838b21cd89d5958fb4b8424d1cb01cddedb328ef69f95c0d3ef46751dea2a
+size 4449793120

models/text_encoder_int8_6bit.json

@@ -1,14 +1,36 @@
 {
-  "model_hash": "003277dd4ed448237a756b6502a65d57",
+  "model_hash": "08ba3e9e8267bc600b0285356c2d5b86",
   "model_name": "moss_image_text_encoder",
   "model_class": "diffsynth.models.moss_image_text_encoder.MOSSImageTextEncoder",
   "quant_config": {
-    "method": "entropack_lossy_quant_int8",
     "load_prequantized": true,
-    "backend_config_kwargs": {
-      "target_bpp": 6.0,
-      "row_rdo_iterations": 2,
-      "row_rdo_candidates": 20
-    }
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 5.956,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "linear_attn.in_proj_a",
+          "linear_attn.in_proj_b"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "linear_attn.in_proj_a",
+          "linear_attn.in_proj_b"
+        ]
+      }
+    ]
   }
 }

models/text_encoder_int8_6bit.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa98d23b48c3af573d441cc027262ec7656759596b671c7d44ed6d469a0ea4d9
-size 7290452872
+oid sha256:040ddf9ab3df1c1363498ea73b67b4dc979a7d843404266e3ad35b0bdc5c2b24
+size 7225461096

models/text_encoder_int8_7bit.json

@@ -0,0 +1,36 @@
+{
+  "model_hash": "1f0b627389760b03f597e10af0e1f4f6",
+  "model_name": "moss_image_text_encoder",
+  "model_class": "diffsynth.models.moss_image_text_encoder.MOSSImageTextEncoder",
+  "quant_config": {
+    "load_prequantized": true,
+    "configs": [
+      {
+        "method": "entropack_lossy_quant_int8",
+        "backend_config_kwargs": {
+          "target_bpp": 6.951,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20,
+          "use_convrot": true,
+          "convrot_groupsize": 256
+        },
+        "exclude_modules": [
+          "linear_attn.in_proj_a",
+          "linear_attn.in_proj_b"
+        ]
+      },
+      {
+        "method": "entropack_lossy_quant",
+        "backend_config_kwargs": {
+          "target_bpp": 8.5,
+          "row_rdo_iterations": 2,
+          "row_rdo_candidates": 20
+        },
+        "target_modules": [
+          "linear_attn.in_proj_a",
+          "linear_attn.in_proj_b"
+        ]
+      }
+    ]
+  }
+}

models/text_encoder_int8_7bit.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf3ee3ff11c380c1ea246f2ed5f06a73cd075686f303ac7a4abe75f3414936c4
+size 8084231096