yestinl commited on
Commit
eec678f
·
1 Parent(s): 27e321a

update readme

Browse files
Files changed (2) hide show
  1. README.md +8 -1
  2. README_CN.md +24 -16
README.md CHANGED
@@ -58,9 +58,16 @@ This repo contains PyTorch model definitions, pretrained weights and inference/s
58
  - [Contents](#contents)
59
  - [Abstract](#abstract)
60
  - [HunyuanImage-2.1 Overall Pipeline](#hunyuanimage-21-overall-pipeline)
 
 
 
 
 
61
  - [🎉 HunyuanImage-2.1 Key Features](#-hunyuanimage-21-key-features)
62
  - [Prompt Enhanced Demo](#prompt-enhanced-demo)
63
  - [📈 Comparisons](#-comparisons)
 
 
64
  - [📜 System Requirements](#-system-requirements)
65
  - [🛠️ Dependencies and Installation](#️-dependencies-and-installation)
66
  - [🧱 Download Pretrained Models](#-download-pretrained-models)
@@ -231,7 +238,7 @@ image = pipe(
231
  width=2048,
232
  height=2048,
233
  use_reprompt=True, # Enable prompt enhancement
234
- use_refiner=False, # Refiner is not supported yet; coming soon.
235
  # For the distilled model, use 8 steps for faster inference.
236
  # For the non-distilled model, use 50 steps for better quality
237
  num_inference_steps=8 if "distilled" in model_name else 50,
 
58
  - [Contents](#contents)
59
  - [Abstract](#abstract)
60
  - [HunyuanImage-2.1 Overall Pipeline](#hunyuanimage-21-overall-pipeline)
61
+ - [Training Data and Caption](#training-data-and-caption)
62
+ - [Text-to-Image Model Architecture](#text-to-image-model-architecture)
63
+ - [Reinforcement Learning from Human Feedback](#reinforcement-learning-from-human-feedback)
64
+ - [Rewriting Model](#rewriting-model)
65
+ - [Model distillation](#model-distillation)
66
  - [🎉 HunyuanImage-2.1 Key Features](#-hunyuanimage-21-key-features)
67
  - [Prompt Enhanced Demo](#prompt-enhanced-demo)
68
  - [📈 Comparisons](#-comparisons)
69
+ - [SSAE Evaluation](#ssae-evaluation)
70
+ - [GSB Evaluation](#gsb-evaluation)
71
  - [📜 System Requirements](#-system-requirements)
72
  - [🛠️ Dependencies and Installation](#️-dependencies-and-installation)
73
  - [🧱 Download Pretrained Models](#-download-pretrained-models)
 
238
  width=2048,
239
  height=2048,
240
  use_reprompt=True, # Enable prompt enhancement
241
+ use_refiner=False, # Disable refiner model, enable it to achieve higher image quality.
242
  # For the distilled model, use 8 steps for faster inference.
243
  # For the non-distilled model, use 50 steps for better quality
244
  num_inference_steps=8 if "distilled" in model_name else 50,
README_CN.md CHANGED
@@ -32,21 +32,29 @@
32
 
33
 
34
  ## 目录
35
- - [🔥🔥🔥 最新动态](#-最新动态)
36
- - [🎥 示例](#-示例)
37
- - [目录](#目录)
38
- - [摘要](#摘要)
39
- - [混元图像 2.1 整体流程](#混元图像-21-整体流程)
40
- - [🎉 关键特性](#-关键特性)
41
- - [提示词增强演示](#提示词增强示例)
42
- - [📈 对比](#-对比)
43
- - [📜 系统要求](#-系统要求)
44
- - [🛠️ 依赖与安装](#-依赖与安装)
45
- - [🧱 模型下载](#-模型下载)
46
- - [🔑 使用](#-使用)
47
- - [🔗 BibTeX](#-bibtex)
48
- - [致谢](#致谢)
49
- - [Github Star 历史](#github-star-历史)
 
 
 
 
 
 
 
 
50
 
51
  ---
52
 
@@ -203,7 +211,7 @@ image = pipe(
203
  width=2048,
204
  height=2048,
205
  use_reprompt=True, # 启用提示词增强
206
- use_refiner=False, # 精修(Refiner)功能暂不支持,后续上线
207
  # 对于蒸馏版模型,建议使用 8 步以加快推理速度;
208
  # 对于非蒸馏版模型,建议使用 50 步以获得更高画质
209
  num_inference_steps=8 if "distilled" in model_name else 50,
 
32
 
33
 
34
  ## 目录
35
+ - [混元图像 2.1:一种用于高分辨率(2K)文本到图像生成的高效扩散模型](#混元图像-21一种用于高分辨率2k文本到图像生成的高效扩散模型)
36
+ - [🔥🔥🔥 最新动态](#-最新动态)
37
+ - [🎥 示例](#-示例)
38
+ - [目录](#目录)
39
+ - [摘要](#摘要)
40
+ - [混元图像 2.1 整体流程](#混元图像-21-整体流程)
41
+ - [训练数据和标注](#训练数据和标注)
42
+ - [文本到图像模型架构](#文本到图像模型架构)
43
+ - [人类反馈强化学习](#人类反馈强化学习)
44
+ - [改写模型](#改写模型)
45
+ - [模型蒸馏](#模型蒸馏)
46
+ - [🎉 关键特性](#-关键特性)
47
+ - [提示词增强示例](#提示词增强示例)
48
+ - [📈 对比](#-对比)
49
+ - [SSAE 评测](#ssae-评测)
50
+ - [GSB 评测](#gsb-评测)
51
+ - [📜 系统要求](#-系统要求)
52
+ - [🛠️ 依赖与安装](#️-依赖与安装)
53
+ - [🧱 模型下载](#-模型下载)
54
+ - [🔑 使用](#-使用)
55
+ - [🔗 BibTeX](#-bibtex)
56
+ - [致谢](#致谢)
57
+ - [Github Star 历史](#github-star-历史)
58
 
59
  ---
60
 
 
211
  width=2048,
212
  height=2048,
213
  use_reprompt=True, # 启用提示词增强
214
+ use_refiner=False, # 关闭精修模型, 可开启以获得更高画质
215
  # 对于蒸馏版模型,建议使用 8 步以加快推理速度;
216
  # 对于非蒸馏版模型,建议使用 50 步以获得更高画质
217
  num_inference_steps=8 if "distilled" in model_name else 50,