yestinl
commited on
Commit
·
eec678f
1
Parent(s):
27e321a
update readme
Browse files- README.md +8 -1
- 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, #
|
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 |
-
- [
|
36 |
-
- [
|
37 |
-
- [
|
38 |
-
- [
|
39 |
-
- [
|
40 |
-
- [
|
41 |
-
- [
|
42 |
-
- [
|
43 |
-
- [
|
44 |
-
- [
|
45 |
-
- [
|
46 |
-
- [
|
47 |
-
- [
|
48 |
-
- [
|
49 |
-
- [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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, #
|
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,
|