sdiazlor commited on
Commit
081f930
·
1 Parent(s): 44a0fb7

Update naming and assets

Browse files
.gitattributes CHANGED
@@ -34,5 +34,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  assets/multi_image.jpg filter=lfs diff=lfs merge=lfs -text
37
- assets/p-h3-image-comfy.jpg filter=lfs diff=lfs merge=lfs -text
38
  assets/quantization.jpg filter=lfs diff=lfs merge=lfs -text
 
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  assets/multi_image.jpg filter=lfs diff=lfs merge=lfs -text
37
+ assets/p-image-oss-comfy.jpg filter=lfs diff=lfs merge=lfs -text
38
  assets/quantization.jpg filter=lfs diff=lfs merge=lfs -text
NOTICE CHANGED
@@ -1,14 +1,14 @@
1
- p-h3-image for ComfyUI
2
  Copyright 2026 Pruna AI
3
 
4
  The files in this repository were modified by Pruna AI.
5
 
6
  diffusion_models/, loras/, vae/ and text_encoders/p-h3-image_text_encoder_*.safetensors
7
- p-h3-image, based on MiniMax-H3 by MiniMax (https://huggingface.co/MiniMaxAI/MiniMax-H3), under the MiniMax H3
8
  Community License Agreement, see LICENSE. The text encoder is also based on Qwen3-VL-4B by the Qwen team
9
  (https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct); its Qwen3-VL-4B parts are under the Apache License 2.0, see
10
  LICENSE-APACHE.
11
 
12
  text_encoders/p-h3-image_prompt_upsampler_*.safetensors
13
- The p-h3-image prompt upsampler, based on Qwen3.5-9B by the Qwen team (https://huggingface.co/Qwen/Qwen3.5-9B),
14
  under the Apache License 2.0, see LICENSE-APACHE.
 
1
+ p-image-oss for ComfyUI
2
  Copyright 2026 Pruna AI
3
 
4
  The files in this repository were modified by Pruna AI.
5
 
6
  diffusion_models/, loras/, vae/ and text_encoders/p-h3-image_text_encoder_*.safetensors
7
+ p-image-oss, based on MiniMax-H3 by MiniMax (https://huggingface.co/MiniMaxAI/MiniMax-H3), under the MiniMax H3
8
  Community License Agreement, see LICENSE. The text encoder is also based on Qwen3-VL-4B by the Qwen team
9
  (https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct); its Qwen3-VL-4B parts are under the Apache License 2.0, see
10
  LICENSE-APACHE.
11
 
12
  text_encoders/p-h3-image_prompt_upsampler_*.safetensors
13
+ The p-image-oss prompt upsampler, based on Qwen3.5-9B by the Qwen team (https://huggingface.co/Qwen/Qwen3.5-9B),
14
  under the Apache License 2.0, see LICENSE-APACHE.
README.md CHANGED
@@ -2,7 +2,7 @@
2
  license: other
3
  license_name: minimax-h3-community-license-agreement
4
  license_link: LICENSE
5
- base_model: PrunaAI/p-h3-image
6
  base_model_relation: quantized
7
  pipeline_tag: text-to-image
8
  language:
@@ -23,7 +23,7 @@ tags:
23
  ---
24
  <!-- header start -->
25
  <p align="center">
26
- <img src="assets/p-h3-image-comfy.jpg" width="800">
27
  </p>
28
  <!-- header end -->
29
 
@@ -50,9 +50,9 @@ tags:
50
  </p>
51
  <div align="center">
52
 
53
- <h1 style="color: #9334E9;">P-H3-Image</h1>
54
 
55
- <h2>ComfyUI files for P-H3-Image, based on MiniMax-H3</h2>
56
 
57
  <h3>
58
  <span style="color: #9334E9;">bf16, int8 and w4a8</span>
@@ -66,18 +66,18 @@ tags:
66
 
67
  </div>
68
 
69
- This repository holds the ComfyUI files for **P-H3-Image**, an image model for text-to-image generation and image
70
  editing based on [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3). It renders 8-megapixel images in 8 steps
71
  without CFG and edits with up to 8 input images. Every component comes in bf16 and in ComfyUI's native int8 and 4-bit
72
  formats. The smallest set is a 20 GB download that peaks at 21 GB of VRAM at 8 MP, against 143 GB of weights for
73
  MiniMax-H3.
74
 
75
- ComfyUI supports P-H3-Image natively, no custom nodes needed. The diffusers pipeline is
76
- in [PrunaAI/P-H3-Image](https://huggingface.co/PrunaAI/P-H3-Image). Try it in the [Hugging Face Spaces](https://huggingface.co/spaces/PrunaAI/P-H3-Image). Read the full blog post [here](https://pruna.ai/blog/oss-p-h3-image-launch).
77
 
78
  ## Quickstart
79
 
80
- 1. **Update ComfyUI.** P-H3-Image support is on ComfyUI's master branch and ships with the next release. Until then,
81
  update to the latest ComfyUI from GitHub (nightly). The desktop app and Comfy Cloud follow stable releases and get it
82
  with the next one.
83
 
@@ -89,13 +89,13 @@ in [PrunaAI/P-H3-Image](https://huggingface.co/PrunaAI/P-H3-Image). Try it in th
89
  peaks at about 56 GB of VRAM:
90
 
91
  ```bash
92
- hf download PrunaAI/P-H3-Image-comfy --include "diffusion_models/p-h3-image_bf16.safetensors" "text_encoders/p-h3-image_text_encoder_bf16.safetensors" "text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors" "vae/p-h3-image_vae_bf16.safetensors" --local-dir ComfyUI/models
93
  ```
94
 
95
  The smallest set, a 20 GB download plus the prompt upsampler, peaks at about 21 GB of VRAM:
96
 
97
  ```bash
98
- hf download PrunaAI/P-H3-Image-Comfy --include "diffusion_models/p-h3-image_w4a8.safetensors" "text_encoders/p-h3-image_text_encoder_w4a8.safetensors" "text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors" "vae/p-h3-image_vae_int8_convrot.safetensors" --local-dir ComfyUI/models
99
  ```
100
 
101
  3. **Open a workflow** from `workflows/`, type a short request and run. The
@@ -148,10 +148,10 @@ NVIDIA Ampere or newer and AMD RDNA3/4.
148
 
149
  | Workflow | What it does |
150
  |---|---|
151
- | `p-h3-image_text_to_image.json` | Text-to-image at 8 MP from a short request |
152
- | `p-h3-image_edit.json` | Edits an image, with optional reference images |
153
- | `p-h3-image_lora_text_to_image.json` | Text-to-image with the LoRA on your own MiniMax-H3 |
154
- | `p-h3-image_lora_edit.json` | Editing with the LoRA on your own MiniMax-H3 |
155
 
156
  The workflows use only ComfyUI's own nodes. Load Diffusion Model loads the model, followed by ModelSamplingMiniMaxH3 at
157
  1.0 for both shifts. Load CLIP (type `minimax`) loads the text encoder, Load VAE the VAE, and p-h3-image Conditioning
@@ -182,7 +182,7 @@ set. Reference images above about 8 MP are scaled down.
182
 
183
  ## Prompt upsampler
184
 
185
- P-H3-Image is sensitive to prompting and renders full, specific prompts much better than short requests. The prompt
186
  upsampler, based on Qwen3.5-9B, turns a one-line request into that prompt, and all example workflows use it. It thinks
187
  about the request first and then writes the prompt.
188
 
@@ -196,7 +196,7 @@ Text, with greedy decoding, so a request always gives the same prompt.
196
  prompt for p-h3-image Conditioning.
197
  - **Use prompt upsampler**, an If/Else switch, set to false sends the request to the model as it is.
198
 
199
- The upsampler runs before P-H3-Image, and ComfyUI moves it out of VRAM when P-H3-Image needs the memory. Its output is
200
  cached, so it only runs again when the request or the images change.
201
 
202
  | Upsampler file | Size | Peak VRAM | Time per prompt, L40S |
@@ -217,13 +217,13 @@ however you like, such as "the hat from the second photo".
217
  <img src="assets/multi_image.jpg" width="100%">
218
  </p>
219
 
220
- *Top, eight public-domain portraits and the dinner party P-H3-Image composed from them, from a one-sentence request.
221
  Bottom, eight scientists at a poker table, seven presidents wading through the swamp with an astronaut, and characters
222
  from our own renders in the same swamp.*
223
 
224
  ## Already have MiniMax-H3?
225
 
226
- If you run MiniMax-H3 in ComfyUI, the 2 GB LoRA `loras/p-h3-image_lora_bf16.safetensors` runs P-H3-Image on its FL2VA
227
  transformer without a second large download. Load `minimax_h3_fl2va_bf16.safetensors` from
228
  [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3) (the pruned and quantized FL2VA files work too),
229
  apply the LoRA with Load LoRA (Bypass, Model Only) at strength 1.0, and start from the `lora` workflows.
@@ -239,7 +239,7 @@ apply the LoRA with Load LoRA (Bypass, Model Only) at strength 1.0, and start fr
239
 
240
  ## License
241
 
242
- The P-H3-Image weights are based on MiniMax-H3 and released under the [MiniMax H3 Community License](LICENSE). Please
243
  read it before use. Among other terms, it excludes the European Union, the United Kingdom, the Republic of Korea and
244
  the United States unless you have a separate license from MiniMax, and products built on it must show "Powered by
245
  MiniMax H3". The text encoder's Qwen3-VL-4B parts and the prompt upsampler, based on Qwen3.5-9B, are under the
 
2
  license: other
3
  license_name: minimax-h3-community-license-agreement
4
  license_link: LICENSE
5
+ base_model: PrunaAI/P-Image-OSS
6
  base_model_relation: quantized
7
  pipeline_tag: text-to-image
8
  language:
 
23
  ---
24
  <!-- header start -->
25
  <p align="center">
26
+ <img src="assets/p-image-oss-comfy.jpg" width="800">
27
  </p>
28
  <!-- header end -->
29
 
 
50
  </p>
51
  <div align="center">
52
 
53
+ <h1 style="color: #9334E9;">P-Image-OSS</h1>
54
 
55
+ <h2>ComfyUI files for P-Image-OSS, based on MiniMax-H3</h2>
56
 
57
  <h3>
58
  <span style="color: #9334E9;">bf16, int8 and w4a8</span>
 
66
 
67
  </div>
68
 
69
+ This repository holds the ComfyUI files for **P-Image-OSS**, an image model for text-to-image generation and image
70
  editing based on [MiniMax-H3](https://huggingface.co/MiniMaxAI/MiniMax-H3). It renders 8-megapixel images in 8 steps
71
  without CFG and edits with up to 8 input images. Every component comes in bf16 and in ComfyUI's native int8 and 4-bit
72
  formats. The smallest set is a 20 GB download that peaks at 21 GB of VRAM at 8 MP, against 143 GB of weights for
73
  MiniMax-H3.
74
 
75
+ ComfyUI supports P-Image-OSS natively, no custom nodes needed. The diffusers pipeline is
76
+ in [PrunaAI/P-Image-OSS](https://huggingface.co/PrunaAI/P-Image-OSS). Try it in the [Hugging Face Spaces](https://huggingface.co/spaces/PrunaAI/P-Image-OSS). Read the full blog post [here](https://pruna.ai/blog/oss-p-h3-image-launch).
77
 
78
  ## Quickstart
79
 
80
+ 1. **Update ComfyUI.** P-Image-OSS support is on ComfyUI's master branch and ships with the next release. Until then,
81
  update to the latest ComfyUI from GitHub (nightly). The desktop app and Comfy Cloud follow stable releases and get it
82
  with the next one.
83
 
 
89
  peaks at about 56 GB of VRAM:
90
 
91
  ```bash
92
+ hf download PrunaAI/P-Image-OSS-Comfy --include "diffusion_models/p-h3-image_bf16.safetensors" "text_encoders/p-h3-image_text_encoder_bf16.safetensors" "text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors" "vae/p-h3-image_vae_bf16.safetensors" --local-dir ComfyUI/models
93
  ```
94
 
95
  The smallest set, a 20 GB download plus the prompt upsampler, peaks at about 21 GB of VRAM:
96
 
97
  ```bash
98
+ hf download PrunaAI/P-Image-OSS-Comfy --include "diffusion_models/p-h3-image_w4a8.safetensors" "text_encoders/p-h3-image_text_encoder_w4a8.safetensors" "text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors" "vae/p-h3-image_vae_int8_convrot.safetensors" --local-dir ComfyUI/models
99
  ```
100
 
101
  3. **Open a workflow** from `workflows/`, type a short request and run. The
 
148
 
149
  | Workflow | What it does |
150
  |---|---|
151
+ | `p-image-oss_text_to_image.json` | Text-to-image at 8 MP from a short request |
152
+ | `p-image-oss_edit.json` | Edits an image, with optional reference images |
153
+ | `p-image-oss_lora_text_to_image.json` | Text-to-image with the LoRA on your own MiniMax-H3 |
154
+ | `p-image-oss_lora_edit.json` | Editing with the LoRA on your own MiniMax-H3 |
155
 
156
  The workflows use only ComfyUI's own nodes. Load Diffusion Model loads the model, followed by ModelSamplingMiniMaxH3 at
157
  1.0 for both shifts. Load CLIP (type `minimax`) loads the text encoder, Load VAE the VAE, and p-h3-image Conditioning
 
182
 
183
  ## Prompt upsampler
184
 
185
+ P-Image-OSS is sensitive to prompting and renders full, specific prompts much better than short requests. The prompt
186
  upsampler, based on Qwen3.5-9B, turns a one-line request into that prompt, and all example workflows use it. It thinks
187
  about the request first and then writes the prompt.
188
 
 
196
  prompt for p-h3-image Conditioning.
197
  - **Use prompt upsampler**, an If/Else switch, set to false sends the request to the model as it is.
198
 
199
+ The upsampler runs before P-Image-OSS, and ComfyUI moves it out of VRAM when P-Image-OSS needs the memory. Its output is
200
  cached, so it only runs again when the request or the images change.
201
 
202
  | Upsampler file | Size | Peak VRAM | Time per prompt, L40S |
 
217
  <img src="assets/multi_image.jpg" width="100%">
218
  </p>
219
 
220
+ *Top, eight public-domain portraits and the dinner party P-Image-OSS composed from them, from a one-sentence request.
221
  Bottom, eight scientists at a poker table, seven presidents wading through the swamp with an astronaut, and characters
222
  from our own renders in the same swamp.*
223
 
224
  ## Already have MiniMax-H3?
225
 
226
+ If you run MiniMax-H3 in ComfyUI, the 2 GB LoRA `loras/p-h3-image_lora_bf16.safetensors` runs P-Image-OSS on its FL2VA
227
  transformer without a second large download. Load `minimax_h3_fl2va_bf16.safetensors` from
228
  [Comfy-Org/MiniMax-H3](https://huggingface.co/Comfy-Org/MiniMax-H3) (the pruned and quantized FL2VA files work too),
229
  apply the LoRA with Load LoRA (Bypass, Model Only) at strength 1.0, and start from the `lora` workflows.
 
239
 
240
  ## License
241
 
242
+ The P-Image-OSS weights are based on MiniMax-H3 and released under the [MiniMax H3 Community License](LICENSE). Please
243
  read it before use. Among other terms, it excludes the European Union, the United Kingdom, the Republic of Korea and
244
  the United States unless you have a separate license from MiniMax, and products built on it must show "Powered by
245
  MiniMax H3". The text encoder's Qwen3-VL-4B parts and the prompt upsampler, based on Qwen3.5-9B, are under the
assets/{p-h3-image-comfy.jpg → p-image-oss-comfy.jpg} RENAMED
File without changes
assets/weights.png CHANGED
workflows/{p-h3-image_edit.json → p-image-oss_edit.json} RENAMED
@@ -480,10 +480,10 @@
480
  "ver": "0.43.0"
481
  },
482
  "widgets_values": [
483
- "p-h3-image"
484
  ],
485
  "widgets_values_named": {
486
- "filename_prefix": "p-h3-image"
487
  }
488
  },
489
  {
@@ -540,7 +540,7 @@
540
  ]
541
  }
542
  ],
543
- "title": "p-h3-image prompt upsampler",
544
  "properties": {
545
  "Node name for S&R": "CLIPLoader",
546
  "cnr_id": "comfy-core",
@@ -548,7 +548,7 @@
548
  "models": [
549
  {
550
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
551
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
552
  "directory": "text_encoders"
553
  }
554
  ]
@@ -1251,7 +1251,7 @@
1251
  "models": [
1252
  {
1253
  "name": "p-h3-image_bf16.safetensors",
1254
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/diffusion_models/p-h3-image_bf16.safetensors",
1255
  "directory": "diffusion_models"
1256
  }
1257
  ]
@@ -1381,13 +1381,13 @@
1381
  ]
1382
  }
1383
  ],
1384
- "title": "p-h3-image text encoder",
1385
  "properties": {
1386
  "Node name for S&R": "CLIPLoader",
1387
  "models": [
1388
  {
1389
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1390
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1391
  "directory": "text_encoders"
1392
  }
1393
  ]
@@ -1444,7 +1444,7 @@
1444
  "models": [
1445
  {
1446
  "name": "p-h3-image_vae_bf16.safetensors",
1447
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1448
  "directory": "vae"
1449
  }
1450
  ]
 
480
  "ver": "0.43.0"
481
  },
482
  "widgets_values": [
483
+ "p-image-oss"
484
  ],
485
  "widgets_values_named": {
486
+ "filename_prefix": "p-image-oss"
487
  }
488
  },
489
  {
 
540
  ]
541
  }
542
  ],
543
+ "title": "p-image-oss prompt upsampler",
544
  "properties": {
545
  "Node name for S&R": "CLIPLoader",
546
  "cnr_id": "comfy-core",
 
548
  "models": [
549
  {
550
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
551
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
552
  "directory": "text_encoders"
553
  }
554
  ]
 
1251
  "models": [
1252
  {
1253
  "name": "p-h3-image_bf16.safetensors",
1254
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/diffusion_models/p-h3-image_bf16.safetensors",
1255
  "directory": "diffusion_models"
1256
  }
1257
  ]
 
1381
  ]
1382
  }
1383
  ],
1384
+ "title": "p-image-oss text encoder",
1385
  "properties": {
1386
  "Node name for S&R": "CLIPLoader",
1387
  "models": [
1388
  {
1389
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1390
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1391
  "directory": "text_encoders"
1392
  }
1393
  ]
 
1444
  "models": [
1445
  {
1446
  "name": "p-h3-image_vae_bf16.safetensors",
1447
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1448
  "directory": "vae"
1449
  }
1450
  ]
workflows/{p-h3-image_lora_edit.json → p-image-oss_lora_edit.json} RENAMED
@@ -545,10 +545,10 @@
545
  "ver": "0.43.0"
546
  },
547
  "widgets_values": [
548
- "p-h3-image"
549
  ],
550
  "widgets_values_named": {
551
- "filename_prefix": "p-h3-image"
552
  }
553
  },
554
  {
@@ -608,7 +608,7 @@
608
  "models": [
609
  {
610
  "name": "p-h3-image_lora_bf16.safetensors",
611
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/loras/p-h3-image_lora_bf16.safetensors",
612
  "directory": "loras"
613
  }
614
  ]
@@ -740,7 +740,7 @@
740
  ]
741
  }
742
  ],
743
- "title": "p-h3-image prompt upsampler",
744
  "properties": {
745
  "Node name for S&R": "CLIPLoader",
746
  "cnr_id": "comfy-core",
@@ -748,7 +748,7 @@
748
  "models": [
749
  {
750
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
751
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
752
  "directory": "text_encoders"
753
  }
754
  ]
@@ -1456,13 +1456,13 @@
1456
  ]
1457
  }
1458
  ],
1459
- "title": "p-h3-image text encoder",
1460
  "properties": {
1461
  "Node name for S&R": "CLIPLoader",
1462
  "models": [
1463
  {
1464
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1465
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1466
  "directory": "text_encoders"
1467
  }
1468
  ]
@@ -1519,7 +1519,7 @@
1519
  "models": [
1520
  {
1521
  "name": "p-h3-image_vae_bf16.safetensors",
1522
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1523
  "directory": "vae"
1524
  }
1525
  ]
 
545
  "ver": "0.43.0"
546
  },
547
  "widgets_values": [
548
+ "p-image-oss"
549
  ],
550
  "widgets_values_named": {
551
+ "filename_prefix": "p-image-oss"
552
  }
553
  },
554
  {
 
608
  "models": [
609
  {
610
  "name": "p-h3-image_lora_bf16.safetensors",
611
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/loras/p-h3-image_lora_bf16.safetensors",
612
  "directory": "loras"
613
  }
614
  ]
 
740
  ]
741
  }
742
  ],
743
+ "title": "p-image-oss prompt upsampler",
744
  "properties": {
745
  "Node name for S&R": "CLIPLoader",
746
  "cnr_id": "comfy-core",
 
748
  "models": [
749
  {
750
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
751
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
752
  "directory": "text_encoders"
753
  }
754
  ]
 
1456
  ]
1457
  }
1458
  ],
1459
+ "title": "p-image-oss text encoder",
1460
  "properties": {
1461
  "Node name for S&R": "CLIPLoader",
1462
  "models": [
1463
  {
1464
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1465
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1466
  "directory": "text_encoders"
1467
  }
1468
  ]
 
1519
  "models": [
1520
  {
1521
  "name": "p-h3-image_vae_bf16.safetensors",
1522
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1523
  "directory": "vae"
1524
  }
1525
  ]
workflows/{p-h3-image_lora_text_to_image.json → p-image-oss_lora_text_to_image.json} RENAMED
@@ -413,10 +413,10 @@
413
  "ver": "0.43.0"
414
  },
415
  "widgets_values": [
416
- "p-h3-image"
417
  ],
418
  "widgets_values_named": {
419
- "filename_prefix": "p-h3-image"
420
  }
421
  },
422
  {
@@ -476,7 +476,7 @@
476
  "models": [
477
  {
478
  "name": "p-h3-image_lora_bf16.safetensors",
479
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/loras/p-h3-image_lora_bf16.safetensors",
480
  "directory": "loras"
481
  }
482
  ]
@@ -608,7 +608,7 @@
608
  ]
609
  }
610
  ],
611
- "title": "p-h3-image prompt upsampler",
612
  "properties": {
613
  "Node name for S&R": "CLIPLoader",
614
  "cnr_id": "comfy-core",
@@ -616,7 +616,7 @@
616
  "models": [
617
  {
618
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
619
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
620
  "directory": "text_encoders"
621
  }
622
  ]
@@ -1149,13 +1149,13 @@
1149
  ]
1150
  }
1151
  ],
1152
- "title": "p-h3-image text encoder",
1153
  "properties": {
1154
  "Node name for S&R": "CLIPLoader",
1155
  "models": [
1156
  {
1157
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1158
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1159
  "directory": "text_encoders"
1160
  }
1161
  ]
@@ -1211,7 +1211,7 @@
1211
  "models": [
1212
  {
1213
  "name": "p-h3-image_vae_bf16.safetensors",
1214
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1215
  "directory": "vae"
1216
  }
1217
  ]
 
413
  "ver": "0.43.0"
414
  },
415
  "widgets_values": [
416
+ "p-image-oss"
417
  ],
418
  "widgets_values_named": {
419
+ "filename_prefix": "p-image-oss"
420
  }
421
  },
422
  {
 
476
  "models": [
477
  {
478
  "name": "p-h3-image_lora_bf16.safetensors",
479
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/loras/p-h3-image_lora_bf16.safetensors",
480
  "directory": "loras"
481
  }
482
  ]
 
608
  ]
609
  }
610
  ],
611
+ "title": "p-image-oss prompt upsampler",
612
  "properties": {
613
  "Node name for S&R": "CLIPLoader",
614
  "cnr_id": "comfy-core",
 
616
  "models": [
617
  {
618
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
619
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
620
  "directory": "text_encoders"
621
  }
622
  ]
 
1149
  ]
1150
  }
1151
  ],
1152
+ "title": "p-image-oss text encoder",
1153
  "properties": {
1154
  "Node name for S&R": "CLIPLoader",
1155
  "models": [
1156
  {
1157
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1158
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1159
  "directory": "text_encoders"
1160
  }
1161
  ]
 
1211
  "models": [
1212
  {
1213
  "name": "p-h3-image_vae_bf16.safetensors",
1214
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1215
  "directory": "vae"
1216
  }
1217
  ]
workflows/{p-h3-image_text_to_image.json → p-image-oss_text_to_image.json} RENAMED
@@ -348,10 +348,10 @@
348
  "ver": "0.43.0"
349
  },
350
  "widgets_values": [
351
- "p-h3-image"
352
  ],
353
  "widgets_values_named": {
354
- "filename_prefix": "p-h3-image"
355
  }
356
  },
357
  {
@@ -408,7 +408,7 @@
408
  ]
409
  }
410
  ],
411
- "title": "p-h3-image prompt upsampler",
412
  "properties": {
413
  "Node name for S&R": "CLIPLoader",
414
  "cnr_id": "comfy-core",
@@ -416,7 +416,7 @@
416
  "models": [
417
  {
418
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
419
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
420
  "directory": "text_encoders"
421
  }
422
  ]
@@ -944,7 +944,7 @@
944
  "models": [
945
  {
946
  "name": "p-h3-image_bf16.safetensors",
947
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/diffusion_models/p-h3-image_bf16.safetensors",
948
  "directory": "diffusion_models"
949
  }
950
  ]
@@ -1074,13 +1074,13 @@
1074
  ]
1075
  }
1076
  ],
1077
- "title": "p-h3-image text encoder",
1078
  "properties": {
1079
  "Node name for S&R": "CLIPLoader",
1080
  "models": [
1081
  {
1082
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1083
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1084
  "directory": "text_encoders"
1085
  }
1086
  ]
@@ -1136,7 +1136,7 @@
1136
  "models": [
1137
  {
1138
  "name": "p-h3-image_vae_bf16.safetensors",
1139
- "url": "https://huggingface.co/PrunaAI/P-H3-Image-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1140
  "directory": "vae"
1141
  }
1142
  ]
 
348
  "ver": "0.43.0"
349
  },
350
  "widgets_values": [
351
+ "p-image-oss"
352
  ],
353
  "widgets_values_named": {
354
+ "filename_prefix": "p-image-oss"
355
  }
356
  },
357
  {
 
408
  ]
409
  }
410
  ],
411
+ "title": "p-image-oss prompt upsampler",
412
  "properties": {
413
  "Node name for S&R": "CLIPLoader",
414
  "cnr_id": "comfy-core",
 
416
  "models": [
417
  {
418
  "name": "p-h3-image_prompt_upsampler_bf16.safetensors",
419
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_prompt_upsampler_bf16.safetensors",
420
  "directory": "text_encoders"
421
  }
422
  ]
 
944
  "models": [
945
  {
946
  "name": "p-h3-image_bf16.safetensors",
947
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/diffusion_models/p-h3-image_bf16.safetensors",
948
  "directory": "diffusion_models"
949
  }
950
  ]
 
1074
  ]
1075
  }
1076
  ],
1077
+ "title": "p-image-oss text encoder",
1078
  "properties": {
1079
  "Node name for S&R": "CLIPLoader",
1080
  "models": [
1081
  {
1082
  "name": "p-h3-image_text_encoder_bf16.safetensors",
1083
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/text_encoders/p-h3-image_text_encoder_bf16.safetensors",
1084
  "directory": "text_encoders"
1085
  }
1086
  ]
 
1136
  "models": [
1137
  {
1138
  "name": "p-h3-image_vae_bf16.safetensors",
1139
+ "url": "https://huggingface.co/PrunaAI/P-Image-OSS-Comfy/resolve/main/vae/p-h3-image_vae_bf16.safetensors",
1140
  "directory": "vae"
1141
  }
1142
  ]