Hugging Face's logo

janhq
/
Jan-v2-VL-med

Image-Text-to-Text
Transformers
Safetensors
English
qwen3_vl
image-to-text
agent
conversational
Model card Files
xet
Community
1
bachvudinh commited on
Commit
33260cf
·
verified ·
1 Parent(s): 8da71bc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -34,6 +34,10 @@ Tasks where the plan and/or knowledge can be provided up front, and success hing
34
 
35
  ## Model Performance
36
 
 
 
 
 
37
  ![image](https://cdn-uploads.huggingface.co/production/uploads/655e3b59d5c0d3db5359ca3c/q4DzuOjmcZOik2c8ZQSCN.png)
38
 
39
  ![image](https://cdn-uploads.huggingface.co/production/uploads/655e3b59d5c0d3db5359ca3c/JdA1kFh2IEJesQsOAOTrh.png)
 
34
 
35
  ## Model Performance
36
 
37
+ ![image](https://cdn-uploads.huggingface.co/production/uploads/655e3b59d5c0d3db5359ca3c/bruqlcVK87KMQE99JsS0c.png)
38
+
39
+ Compared with its base (**[Qwen-3-VL-8B-Thinking](https://huggingface.co/Qwen/Qwen3-VL-8B-Thinking)**), **Jan-v2-VL** shows **no degradation** on standard text-only and vision tasks—and is **slightly better on several**—while delivering stronger long-horizon execution on the *Illusion of Diminishing Returns* benchmark.
40
+
41
  ![image](https://cdn-uploads.huggingface.co/production/uploads/655e3b59d5c0d3db5359ca3c/q4DzuOjmcZOik2c8ZQSCN.png)
42
 
43
  ![image](https://cdn-uploads.huggingface.co/production/uploads/655e3b59d5c0d3db5359ca3c/JdA1kFh2IEJesQsOAOTrh.png)