Ryan1122 commited on
Commit
699ff5c
·
verified ·
1 Parent(s): 1cf6b00

Preserve latest main license information and model names

Browse files
Files changed (1) hide show
  1. README.md +16 -6
README.md CHANGED
@@ -1,5 +1,8 @@
1
  ---
2
  pipeline_tag: text-generation
 
 
 
3
  library_name: transformers
4
  language:
5
  - en
@@ -25,6 +28,7 @@ tags:
25
  <a href="https://modelscope.cn/organization/minimax" target="_blank" rel="noopener noreferrer"><img alt="ModelScope MiniMax AI" src="https://img.shields.io/badge/ModelScope-MiniMax%20AI-white?labelColor=%23EF3D5D"></a>
26
  <a href="https://platform.minimaxi.com/docs/faq/contact-us" target="_blank"><img src="https://img.shields.io/badge/WeChat-07C160?logo=wechat&logoColor=white" alt="WeChat"></a>
27
  <a href="https://discord.com/invite/dbMxutw7tP" target="_blank"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Discord"></a>
 
28
  </p>
29
 
30
  # MiniMax-H3-Image-PE
@@ -51,7 +55,7 @@ This repository covers two workflows:
51
 
52
  | Workflow | Input | Released assets |
53
  | --- | --- | --- |
54
- | **Text-to-image (T2I) PE** | A text instruction | Fine-tuned Qwen3.8-27B weights in FP4, task and system prompts, and a client example |
55
  | **Image editing PE** | An editing instruction and reference images | A task prompt and message-assembly example for a vision-language model of your choice |
56
 
57
  T2I PE expands the scene, composition and visual details of a generation request.
@@ -59,21 +63,21 @@ Image editing PE describes the intended changes while preserving relevant conten
59
  reference images. **Image editing PE is still under development; this release provides its
60
  prompt only.**
61
 
62
- The examples below explain how to use the released T2I model and image editing task prompt.
63
 
64
  ## How to use T2I PE
65
 
66
  Run the examples from a local copy of this repository so the relative prompt and
67
  script paths resolve correctly.
68
 
69
- ### Serve the model
70
 
71
  We recommend SGLang with `sglang>=0.5.23` and `xgrammar>=0.2.7`.
72
 
73
  ```bash
74
  sglang serve \
75
  --model-path MiniMaxAI/MiniMax-H3-Image-PE \
76
- --served-model-name h3-image-pe \
77
  --model-type llm \
78
  --reasoning-parser qwen3 \
79
  --grammar-backend xgrammar \
@@ -108,7 +112,7 @@ system_prompt = Path("prompt/t2i/system_prompt.txt").read_text(encoding="utf-8")
108
  params = json.loads(Path("prompt/t2i/request_params.json").read_text(encoding="utf-8"))
109
  payload = {
110
  **params,
111
- "model": "h3-image-pe",
112
  "messages": [
113
  {"role": "system", "content": system_prompt},
114
  {"role": "user", "content": user_prompt},
@@ -131,7 +135,7 @@ The included client performs the same request and checks the response:
131
  ```bash
132
  python3 scripts/expand_prompt.py \
133
  --url http://0.0.0.0:30000 \
134
- --model h3-image-pe \
135
  "An orange tabby cat sleeping on a sunlit windowsill."
136
  ```
137
 
@@ -250,3 +254,9 @@ and describes new content. Descriptive prose is English; literal text retains th
250
  language you request. `target_aspect_ratio` is a positive-integer `W:H` string.
251
  If you select an explicit ratio, the model preserves it exactly. If you select
252
  `adaptive`, the model determines a numeric ratio for the intended edited image.
 
 
 
 
 
 
 
1
  ---
2
  pipeline_tag: text-generation
3
+ license: other
4
+ license_name: minimax-h3-image-community-license-agreement
5
+ license_link: https://huggingface.co/MiniMaxAI/MiniMax-H3-Image/blob/main/LICENSE
6
  library_name: transformers
7
  language:
8
  - en
 
28
  <a href="https://modelscope.cn/organization/minimax" target="_blank" rel="noopener noreferrer"><img alt="ModelScope MiniMax AI" src="https://img.shields.io/badge/ModelScope-MiniMax%20AI-white?labelColor=%23EF3D5D"></a>
29
  <a href="https://platform.minimaxi.com/docs/faq/contact-us" target="_blank"><img src="https://img.shields.io/badge/WeChat-07C160?logo=wechat&logoColor=white" alt="WeChat"></a>
30
  <a href="https://discord.com/invite/dbMxutw7tP" target="_blank"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Discord"></a>
31
+ <a href="https://huggingface.co/MiniMaxAI/MiniMax-H3-Image/blob/main/LICENSE"><img src="https://img.shields.io/badge/LICENSE-4CAF50?logo=creativecommons&logoColor=white" alt="LICENSE"></a>
32
  </p>
33
 
34
  # MiniMax-H3-Image-PE
 
55
 
56
  | Workflow | Input | Released assets |
57
  | --- | --- | --- |
58
+ | **Text-to-image (T2I) PE** | A text instruction | MiniMax-H3-Image-PE weights (fine-tuned from Qwen3.8-27B, FP4), task and system prompts, and a client example |
59
  | **Image editing PE** | An editing instruction and reference images | A task prompt and message-assembly example for a vision-language model of your choice |
60
 
61
  T2I PE expands the scene, composition and visual details of a generation request.
 
63
  reference images. **Image editing PE is still under development; this release provides its
64
  prompt only.**
65
 
66
+ The examples below explain how to use MiniMax-H3-Image-PE and the image editing task prompt.
67
 
68
  ## How to use T2I PE
69
 
70
  Run the examples from a local copy of this repository so the relative prompt and
71
  script paths resolve correctly.
72
 
73
+ ### Serve MiniMax-H3-Image-PE
74
 
75
  We recommend SGLang with `sglang>=0.5.23` and `xgrammar>=0.2.7`.
76
 
77
  ```bash
78
  sglang serve \
79
  --model-path MiniMaxAI/MiniMax-H3-Image-PE \
80
+ --served-model-name MiniMax-H3-Image-PE \
81
  --model-type llm \
82
  --reasoning-parser qwen3 \
83
  --grammar-backend xgrammar \
 
112
  params = json.loads(Path("prompt/t2i/request_params.json").read_text(encoding="utf-8"))
113
  payload = {
114
  **params,
115
+ "model": "MiniMax-H3-Image-PE",
116
  "messages": [
117
  {"role": "system", "content": system_prompt},
118
  {"role": "user", "content": user_prompt},
 
135
  ```bash
136
  python3 scripts/expand_prompt.py \
137
  --url http://0.0.0.0:30000 \
138
+ --model MiniMax-H3-Image-PE \
139
  "An orange tabby cat sleeping on a sunlit windowsill."
140
  ```
141
 
 
254
  language you request. `target_aspect_ratio` is a positive-integer `W:H` string.
255
  If you select an explicit ratio, the model preserves it exactly. If you select
256
  `adaptive`, the model determines a numeric ratio for the intended edited image.
257
+
258
+ ## License
259
+
260
+ - MiniMax-H3-Image-PE is part of MiniMax-H3-Image and is released under the [MiniMax-H3-Image Community License Agreement](https://huggingface.co/MiniMaxAI/MiniMax-H3-Image/blob/main/LICENSE).
261
+ - MiniMax-H3-Image-PE is derived from [Qwen3.8-27B](https://huggingface.co/Qwen/Qwen3.8-27B), which is licensed under the Apache 2.0 License.
262
+ - The separate written authorization described in Section IV of the license is requested through [platform.minimax.io/license](https://platform.minimax.io/license).