update readme
Browse files- README.md +3 -1
- README_CN.md +3 -1
README.md
CHANGED
@@ -200,11 +200,13 @@ From the results, HunyuanImage 2.1 achieved a relative win rate of -1.36% agains
|
|
200 |
|
201 |
**Hardware and OS Requirements:**
|
202 |
- NVIDIA GPU with CUDA support.
|
203 |
-
|
|
|
204 |
|
205 |
> ✨ An FP8-quantized model is coming soon, enabling even lower GPU memory requirements for inference, stay tuned 👀!
|
206 |
|
207 |
> **Note:** The memory requirements above are measured with model CPU offloading enabled. If your GPU has sufficient memory, you may disable offloading for improved inference speed.
|
|
|
208 |
- Supported operating system: Linux.
|
209 |
|
210 |
|
|
|
200 |
|
201 |
**Hardware and OS Requirements:**
|
202 |
- NVIDIA GPU with CUDA support.
|
203 |
+
|
204 |
+
**Minimum requrement for now:** 36 GB GPU memory for 2048x2048 image generation.
|
205 |
|
206 |
> ✨ An FP8-quantized model is coming soon, enabling even lower GPU memory requirements for inference, stay tuned 👀!
|
207 |
|
208 |
> **Note:** The memory requirements above are measured with model CPU offloading enabled. If your GPU has sufficient memory, you may disable offloading for improved inference speed.
|
209 |
+
|
210 |
- Supported operating system: Linux.
|
211 |
|
212 |
|
README_CN.md
CHANGED
@@ -176,11 +176,13 @@ SSAE(结构化语义对齐评估)是一种基于先进多模态大语言模
|
|
176 |
|
177 |
**硬件和操作系统要求:**
|
178 |
- 支持 CUDA 的 NVIDIA GPU。
|
179 |
-
|
|
|
180 |
|
181 |
> ✨ 即将推出 FP8 量化模型,推理所需显存将进一步降低,敬请期待 👀!
|
182 |
|
183 |
> **注意:** 上述内存要求是在启用模型 CPU offloading 的情况下测量的。如果您的 GPU 有足够的显存,可以禁用 CPU offloading 以提高推理速度。
|
|
|
184 |
- 支持的操作系统:Linux。
|
185 |
|
186 |
|
|
|
176 |
|
177 |
**硬件和操作系统要求:**
|
178 |
- 支持 CUDA 的 NVIDIA GPU。
|
179 |
+
|
180 |
+
**最低要求:** 36 GB 显存,可用于 2048x2048 图像生成。
|
181 |
|
182 |
> ✨ 即将推出 FP8 量化模型,推理所需显存将进一步降低,敬请期待 👀!
|
183 |
|
184 |
> **注意:** 上述内存要求是在启用模型 CPU offloading 的情况下测量的。如果您的 GPU 有足够的显存,可以禁用 CPU offloading 以提高推理速度。
|
185 |
+
|
186 |
- 支持的操作系统:Linux。
|
187 |
|
188 |
|