Halve @spaces.GPU duration 60s → 30s to save quota per call

System prompt: explicit greeting + short-followup rules; never introduce self

Slash system prompt: short, answer directly, no identity preamble

Bulletproof type coercion + full traceback logging in chat_fn

Ignore Gradio session history + bake identity facts into system prompt

Ignore Gradio session history + bake identity facts into system prompt

Defensive history shape + voice-rewrite of system prompt

Fix: handle both dict + tuple history shapes in format_mistral_prompt

Revert gradio 5.x bump (HF Dockerfile forces 4.44); pin starlette+fastapi instead

Revert gradio 5.x bump (HF Dockerfile forces 4.44); pin starlette+fastapi instead

Upgrade to Gradio 5.x (fixes Starlette TemplateResponse signature bug)

Pin jinja2<3.1.5 to work around Gradio 4.x template cache TypeError

Simplify: drop multi-tab Blocks wrapper (broken queue init in Gradio 4.x)

Fix: ChatInterface in Gradio 4.x doesn't accept api_name kwarg

Bootstrap: app.py

Bootstrap: llm.py

Bootstrap: requirements.txt

Bootstrap: README.md

initial commit