Hugging Face's logo

THUDM
/
CogVideoX1.5-5B

Image-to-Video
Diffusers
Safetensors
English
CogVideoXPipeline
video-generation
thudm
Model card Files Community
2
zR commited on
Commit
350b2b0
•
1 Parent(s): e787df2

update

Browse files
Files changed (2) hide show
  1. README.md +5 -6
  2. README_zh.md +5 -6
README.md CHANGED
@@ -48,16 +48,15 @@ Below is a table listing information on the video generation models available in
48
  <td colspan="2" style="text-align: center;"><b>BF16 (recommended)</b>, FP16, FP32, FP8*, INT8, not supported INT4</td>
49
  </tr>
50
  <tr>
51
- <td style="text-align: center;">Single GPU Inference VRAM Consumption</td>
52
- <td style="text-align: center;"><b>BF16: 5GB minimum*</b><br><b>diffusers INT8 (torchao): 4.4GB minimum*</b></td>
53
- <td style="text-align: center;"><b>BF16: 5GB minimum*</b><br><b>diffusers INT8 (torchao): 4.4GB minimum*</b></td>
54
  </tr>
55
  <tr>
56
- <td style="text-align: center;">Multi-GPU Inference VRAM Consumption</td>
57
- <td colspan="2" style="text-align: center;"><b>BF16: 15GB* using diffusers</b><br></td>
58
  </tr>
59
  <tr>
60
- <td style="text-align: center;">Inference Speed<br>(Step = 50, FP/BF16)</td>
61
  <td colspan="2" style="text-align: center;">Single A100: ~1000 seconds (5-second video)<br>Single H100: ~550 seconds (5-second video)</td>
62
  </tr>
63
  <tr>
 
48
  <td colspan="2" style="text-align: center;"><b>BF16 (recommended)</b>, FP16, FP32, FP8*, INT8, not supported INT4</td>
49
  </tr>
50
  <tr>
51
+ <td style="text-align: center;">Single GPU Inference Memory Consumption</td>
52
+ <td colspan="2" style="text-align: center;"><b>BF16: 9GB minimum*</b></td>
 
53
  </tr>
54
  <tr>
55
+ <td style="text-align: center;">Multi-GPU Inference Memory Consumption</td>
56
+ <td colspan="2" style="text-align: center;"><b>BF16: 15GB* </b><br></td>
57
  </tr>
58
  <tr>
59
+ <td style="text-align: center;">Inference Speed<br>(Step = 50, BF16)</td>
60
  <td colspan="2" style="text-align: center;">Single A100: ~1000 seconds (5-second video)<br>Single H100: ~550 seconds (5-second video)</td>
61
  </tr>
62
  <tr>
README_zh.md CHANGED
@@ -34,16 +34,15 @@ CogVideoX是 [清影](https://chatglm.cn/video?fr=osm_cogvideo) 同源的开源
34
  <td colspan="2" style="text-align: center;"><b>BF16(推荐)</b>, FP16, FP32，FP8*，INT8，不支持INT4</td>
35
  </tr>
36
  <tr>
37
- <td style="text-align: center;">单GPU推理显存消耗</td>
38
- <td style="text-align: center;"><b> BF16 : 5GB起* </b><br><b>diffusers INT8(torchao): 4.4G起* </b></td>
39
- <td style="text-align: center;"><b> BF16 : 5GB起* </b><br><b>diffusers INT8(torchao): 4.4G起* </b></td>
40
  </tr>
41
  <tr>
42
- <td style="text-align: center;">多GPU推理显存消耗</td>
43
- <td colspan="2" style="text-align: center;"><b>BF16: 15GB* using diffusers</b><br></td>
44
  </tr>
45
  <tr>
46
- <td style="text-align: center;">推理速度<br>(Step = 50, FP/BF16)</td>
47
  <td colspan="2" style="text-align: center;">单卡A100: ~1000秒(5秒视频)<br>单卡H100: ~550秒(5秒视频)</td>
48
  </tr>
49
  <tr>
 
34
  <td colspan="2" style="text-align: center;"><b>BF16(推荐)</b>, FP16, FP32，FP8*，INT8，不支持INT4</td>
35
  </tr>
36
  <tr>
37
+ <td style="text-align: center;">单GPU显存消耗</td>
38
+ <td colspan="2" style="text-align: center;"><b>BF16: 9GB minimum*</b></td>
 
39
  </tr>
40
  <tr>
41
+ <td style="text-align: center;">多GPU显存消耗</td>
42
+ <td colspan="2" style="text-align: center;"><b>BF16: 15GB* </b><br></td>
43
  </tr>
44
  <tr>
45
+ <td style="text-align: center;">推理速度<br>(Step = 50, BF16)</td>
46
  <td colspan="2" style="text-align: center;">单卡A100: ~1000秒(5秒视频)<br>单卡H100: ~550秒(5秒视频)</td>
47
  </tr>
48
  <tr>