Install diffusers from GitHub and use the new repo name

README.md

@@ -67,11 +67,11 @@ tags:
 **P-H3-Image** is an image model for text-to-image generation and image editing, based on
 [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3). It renders 8-megapixel images in 8 steps without
 classifier-free guidance, edits with up to 8 input images, and comes with a prompt upsampler that turns a one-line
-request into the full prompt p-h3-image renders best. It is strongest at photorealism, film and editorial looks, and
+request into the full prompt P-H3-Image renders best. It is strongest at photorealism, film and editorial looks, and
 graphic styles with short quoted text.
 
 This repository holds the [diffusers](https://github.com/huggingface/diffusers) pipeline. For ComfyUI, with int8 and
-4-bit files that run in 21 GB of VRAM, see [PrunaAI/p-h3-image-comfy](https://huggingface.co/PrunaAI/p-h3-image-comfy).
+4-bit files that run in 21 GB of VRAM, see [PrunaAI/P-H3-Image-Comfy](https://huggingface.co/PrunaAI/P-H3-Image-Comfy).
 
 | | |
 |---|---|
@@ -85,17 +85,18 @@ This repository holds the [diffusers](https://github.com/huggingface/diffusers)
 ## Quickstart
 
 Runs text-to-image and editing with the same pipeline. At 8 MP in bf16 it needs a GPU with about 101 GB of memory,
-such as one H200 or B200.
+such as one H200 or B200, or 80 GB with [offloading](#less-gpu-memory).
 
 ```bash
-pip install -U diffusers transformers accelerate torch pillow
+pip install git+https://github.com/huggingface/diffusers
+pip install -U transformers accelerate torch pillow
 ```
 
 ```python
 import torch
 from diffusers import ModularPipeline
 
-pipe = ModularPipeline.from_pretrained("PrunaAI/p-h3-image")
+pipe = ModularPipeline.from_pretrained("PrunaAI/P-H3-Image")
 pipe.load_components(dtype=torch.bfloat16)
 pipe.to("cuda")
 
@@ -134,7 +135,7 @@ composed = pipe(
   <img src="assets/resolutions.png" width="100%">
 </p>
 
-p-h3-image works best at **8 MP**. Images at 2 and 4 MP can be good too, but artifacts become more frequent, especially
+P-H3-Image works best at **8 MP**. Images at 2 and 4 MP can be good too, but artifacts become more frequent, especially
 in faces, hands and small details. For text-to-image, use one of the 8 MP sizes below. The default is 2912 × 2912.
 
 Edits work best at 4 or 8 MP. Without `height` and `width`, the edit is generated at the size of Picture 1, scaled up to
@@ -176,7 +177,7 @@ resolutions because the noise changes with the image size.*
 
 ## Prompting
 
-p-h3-image is sensitive to prompting and renders full, specific prompts much better than short requests. The pipeline
+P-H3-Image is sensitive to prompting and renders full, specific prompts much better than short requests. The pipeline
 therefore runs a prompt upsampler, based on Qwen3.5-9B, on every request. It writes a detailed prompt for the aspect
 ratio of `height` and `width`, or, for an edit, a brief of what to keep and what to change. The prompt it wrote comes
 back as the `prompt` output.
@@ -189,10 +190,10 @@ print(out["prompt"])
 To render a prompt as written, remove the step from the blocks.
 
 ```python
-blocks = ModularPipeline.from_pretrained("PrunaAI/p-h3-image").blocks
+blocks = ModularPipeline.from_pretrained("PrunaAI/P-H3-Image").blocks
 blocks.sub_blocks.pop("prompt_upsampler")
 
-pipe = blocks.init_pipeline("PrunaAI/p-h3-image")
+pipe = blocks.init_pipeline("PrunaAI/P-H3-Image")
 pipe.load_components(dtype=torch.bfloat16)
 pipe.to("cuda")
 image = pipe(prompt=prompt, height=3360, width=2496, output="images")[0]  # rendered as written
@@ -208,7 +209,7 @@ When you write prompts by hand, four things matter most.
 3. **Spell out every word.** Text without its exact words turns into pseudo-text. Put each piece of text in quotes,
    keep each string short, say where it sits, and end with "no other text". If you want no text, don't mention signs
    or labels at all.
-4. **Leave out hype words.** Stunning, epic, cinematic, vibrant, hyper-detailed, 8k or masterpiece push p-h3-image
+4. **Leave out hype words.** Stunning, epic, cinematic, vibrant, hyper-detailed, 8k or masterpiece push P-H3-Image
    toward a glossy, oversaturated, airbrushed look.
 
 <p align="center">
@@ -231,7 +232,7 @@ text.
 ## Editing
 
 Pass one image to edit it, or a list of up to 8. Picture 1 is the image to edit, and the others are references the
-request can name as Picture 2, Picture 3 and so on. The upsampler writes the edit prompt in the form p-h3-image edits
+request can name as Picture 2, Picture 3 and so on. The upsampler writes the edit prompt in the form P-H3-Image edits
 best from, so a short request such as "make the pickup truck red" is enough.
 
 <p align="center">
@@ -239,17 +240,17 @@ best from, so a short request such as "make the pickup truck red" is enough.
 </p>
 
 *The same short request given as typed (middle) and written as an edit prompt that names what to keep (right), the form
-the upsampler writes. As typed, p-h3-image often redraws the whole scene. As an edit prompt, only the requested part
+the upsampler writes. As typed, P-H3-Image often redraws the whole scene. As an edit prompt, only the requested part
 changes.*
 
-p-h3-image combines many references at once. Given up to 8 pictures and a one-line request, the upsampler names each
-picture and what to take from it, and p-h3-image composes them into one scene.
+P-H3-Image combines many references at once. Given up to 8 pictures and a one-line request, the upsampler names each
+picture and what to take from it, and P-H3-Image composes them into one scene.
 
 <p align="center">
   <img src="assets/multi_image.jpg" width="100%">
 </p>
 
-*Top, eight public-domain portraits and the dinner party p-h3-image composed from them, from a one-sentence request.
+*Top, eight public-domain portraits and the dinner party P-H3-Image composed from them, from a one-sentence request.
 Bottom, eight scientists at a poker table, seven presidents wading through the swamp with an astronaut, and characters
 from our own renders in the same swamp.*
 
@@ -264,9 +265,26 @@ Edit <Picture 1>. <The change, described as the finished result>. Keep <everythi
 - **Describe the result, not the action.** Say how the edited part looks in the finished image, and give new things a
   size next to something already there, such as "about the size of the cat's head".
 - **Say what stays.** List what must not change, such as faces, poses, the background and the light. Without this
-  list, p-h3-image tends to redraw the whole image.
+  list, P-H3-Image tends to redraw the whole image.
 - **Quote new text exactly.** Put new text in quotes and name the text around it that must stay.
 
+## Less GPU memory
+
+With auto offloading, diffusers moves each component onto the GPU when a step needs it, so the pipeline runs on one
+80 GB GPU such as an H100, peaking at about 76 GB at 8 MP.
+
+```python
+import torch
+from diffusers import ComponentsManager, ModularPipeline
+
+manager = ComponentsManager()
+manager.enable_auto_cpu_offload(device="cuda")
+
+pipe = ModularPipeline.from_pretrained("PrunaAI/P-H3-Image", components_manager=manager)
+pipe.load_components(dtype=torch.bfloat16)
+image = pipe(prompt=prompt, height=3360, width=2496, output="images")[0]
+```
+
 ## Performance
 
 One NVIDIA B200, bf16, batch size 1, 8 MP, no compilation or offloading.
@@ -278,7 +296,7 @@ One NVIDIA B200, bf16, batch size 1, 8 MP, no compilation or offloading.
 
 The 8 steps run over the checkpoint's own sampling grid. Pass `sigmas` to use another one. There is no guider,
 `negative_prompt` or `guidance_scale`. For GPUs with less memory, use the int8 or w4a8 files in
-[PrunaAI/p-h3-image-comfy](https://huggingface.co/PrunaAI/p-h3-image-comfy).
+[PrunaAI/P-H3-Image-Comfy](https://huggingface.co/PrunaAI/P-H3-Image-Comfy).
 
 ## Limitations
 
@@ -289,17 +307,17 @@ The 8 steps run over the checkpoint's own sampling grid. Pass `sigmas` to use an
   longer passages sometimes get single letters wrong. Short quoted strings render well.
 - **Lower resolutions show more artifacts.** At 2 and 4 MP, faces, hands and small details break more often than at
   8 MP.
-- **Single images only.** p-h3-image does not generate video or audio.
+- **Single images only.** P-H3-Image does not generate video or audio.
 
 ## License
 
-p-h3-image is based on MiniMax-H3 and released under the [MiniMax H3 Community License](LICENSE). Please read it before
+P-H3-Image is based on MiniMax-H3 and released under the [MiniMax H3 Community License](LICENSE). Please read it before
 use. Among other terms, it excludes the European Union, the United Kingdom, the Republic of Korea and the United States
 unless you have a separate license from MiniMax, and products built on it must show "Powered by MiniMax H3". The text
 encoder's Qwen3-VL-4B parts and the prompt upsampler, based on Qwen3.5-9B, are under the
 [Apache License 2.0](LICENSE-APACHE). See [NOTICE](NOTICE) for details.
 
-p-h3-image would not exist without [MiniMax](https://huggingface.co/MiniMaxAI), who released H3 with open weights, and
+P-H3-Image would not exist without [MiniMax](https://huggingface.co/MiniMaxAI), who released H3 with open weights, and
 the [Qwen team](https://huggingface.co/Qwen).
 
 ## What's next?

modular_model_index.json

@@ -25,7 +25,7 @@
         "transformers",
         "Qwen3VLForConditionalGeneration"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "text_encoder",
       "variant": null,
       "revision": null
@@ -39,7 +39,7 @@
         "diffusers",
         "PH3ImageTextConnector"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "text_connector",
       "variant": null,
       "revision": null
@@ -53,7 +53,7 @@
         "transformers",
         "Qwen2TokenizerFast"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "tokenizer",
       "variant": null,
       "revision": null
@@ -67,7 +67,7 @@
         "transformers",
         "Qwen3VLProcessor"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "processor",
       "variant": null,
       "revision": null
@@ -81,7 +81,7 @@
         "diffusers",
         "AutoencoderKLPH3Image"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "vae",
       "variant": null,
       "revision": null
@@ -95,7 +95,7 @@
         "diffusers",
         "MiniMaxH3Transformer3DModel"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "transformer",
       "variant": null,
       "revision": null
@@ -109,7 +109,7 @@
         "diffusers",
         "MiniMaxH3Scheduler"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "scheduler",
       "variant": null,
       "revision": null
@@ -123,7 +123,7 @@
         "diffusers",
         "MiniMaxH3Scheduler"
       ],
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "subfolder": "audio_scheduler",
       "variant": null,
       "revision": null
@@ -133,7 +133,7 @@
     "transformers",
     "Qwen3_5ForConditionalGeneration",
     {
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "revision": null,
       "subfolder": "prompt_upsampler",
       "type_hint": [
@@ -147,7 +147,7 @@
     "transformers",
     "AutoProcessor",
     {
-      "pretrained_model_name_or_path": "PrunaAI/p-h3-image",
+      "pretrained_model_name_or_path": "PrunaAI/P-H3-Image",
       "revision": null,
       "subfolder": "prompt_upsampler_processor",
       "type_hint": [