Hugging Face's logo

AXERA-TECH
/
SmolVLM2-256M-Video-Instruct_Ax650

Image-Text-to-Text
Transformers
English
Chinese
Int8
VLM
Model card Files
xet
Community
lihongjie commited on
Commit
94489eb
·
1 Parent(s): 9b30095

update

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -36,12 +36,12 @@ For those who are interested in model conversion, you can try to export axmodel
36
  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)
37
 
38
  **Image Process**
39
- |Chips| input size | image num | image encoder | ttft(168 tokens) | w8a16 | CMM | Flash |
40
  |--|--|--|--|--|--|--|--|
41
  |AX650| 512*512 | 1 | 516 ms | 271 ms | 76.7 tokens/sec| 455 MB | 415MB |
42
 
43
  **Video Process**
44
- |Chips| input size | image num | image encoder |ttft(600 tokens) | w8a16 | CMM | Flash |
45
  |--|--|--|--|--|--|--|--|
46
  |AX650| 512*512 | 8 | 834 ms | 827 ms | 75.46 tokens/sec| 455 MB | 415MB |
47
 
 
36
  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)
37
 
38
  **Image Process**
39
+ |Chips| input size | image num | image encoder | ttft(344 tokens) | w8a16 | CMM | Flash |
40
  |--|--|--|--|--|--|--|--|
41
  |AX650| 512*512 | 1 | 516 ms | 271 ms | 76.7 tokens/sec| 455 MB | 415MB |
42
 
43
  **Video Process**
44
+ |Chips| input size | image num | image encoder |ttft(656 tokens) | w8a16 | CMM | Flash |
45
  |--|--|--|--|--|--|--|--|
46
  |AX650| 512*512 | 8 | 834 ms | 827 ms | 75.46 tokens/sec| 455 MB | 415MB |
47