Hugging Face's logo

AXERA-TECH
/
Qwen3-VL-2B-Instruct

Image-Text-to-Text
Transformers
English
Chinese
Qwen3-VL
Qwen3-VL-2B-Instruct
Qwen3-VL-4B-Instruct
Int8
VLM
Model card Files
xet
Community
1
lihongjie commited on
Commit
7c180e2
·
1 Parent(s): 2e38943

更新prefill token num; 加载后处理参数

Browse files
Files changed (2) hide show
  1. README.md +1 -1
  2. main_ax650 +2 -2
README.md CHANGED
@@ -42,7 +42,7 @@ For those who are interested in model conversion, you can try to export axmodel
42
  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)
43
 
44
  **Image Process**
45
- |Chips| input size | image num | image encoder | ttft(320 tokens) | w8a16 | CMM | Flash |
46
  |--|--|--|--|--|--|--|--|
47
  |AX650| 384*384 | 1 | 238 ms | 392 ms | 9.5 tokens/sec| 4.1GiB | 4.2GiB |
48
 
 
42
  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)
43
 
44
  **Image Process**
45
+ |Chips| input size | image num | image encoder | ttft(168 tokens) | w8a16 | CMM | Flash |
46
  |--|--|--|--|--|--|--|--|
47
  |AX650| 384*384 | 1 | 238 ms | 392 ms | 9.5 tokens/sec| 4.1GiB | 4.2GiB |
48
 
main_ax650 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cf155daea2ad2f86890b03acd17d887f5e9a7f5b9c8cf16a1b2c2c7cf4fa2520
3
- size 6598288
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b9b7a1dd91f4b183324d86f537c1530a075bbb9d551cc8fc24d0158e9d513e1
3
+ size 6660400