Pin (profile order 4)

Pin (profile order 4)

Require Sign in with HF: inference always uses the visitor's own token, no Space token or anonymous fallback

Default to Qwen3-VL-30B-A3B-Instruct; add gemma-3-27b-it and Llama-4-Scout

Gradio 5.28->6.29: messages-format chat (images now actually sent with history), OAuth token injected directly, served model list (Qwen2.5-VL-7B default), theme to launch()

Update app.py

Update README.md

Update app.py

Update app.py

Update README.md

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update requirements.txt

Update app.py

Update app.py

Update app.py

Update app.py

adding tool use

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

adding all smolagents addons

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update app.py

Update requirements.txt

testing MCP support

Update app.py

Update README.md

Update app.py