Spaces:
Runtime error
Runtime error
Commit History
-
ULTRA SPEED: 8-bit quantization, greedy decoding, 40 tokens, inference_mode
Ilke Ileric46fe44 -
Optimize for speed: reduce to 80 tokens, lower sampling params, add timing logs
Ilke Ileri190133f -
Improve conversation quality: use full history, increase tokens to 150, better sampling
Ilke Ileri39bb917 -
Add /v1/chat/completions route for Vapi compatibility
Ilke Ileri2311595 -
Optimize for speed: max_tokens=50, greedy decoding for real-time voice
Ilke Ilerib9e9889 -
Add debug endpoint and improve error handling for Vapi troubleshooting
Ilke Ileri88b53d1 -
Reduce max_tokens to 100 for faster response to prevent Vapi timeout
Ilke Ilerid54cae5
-
Add streaming support for Vapi compatibility
Ilke Ileri0ccd1fa -
Fix: remove global counter causing Space crash
Ilke Ileric6b30d3 -
Add request counter to track incoming Vapi requests
Ilke Ileridcb9815 -
Fix: properly extract user messages from Vapi conversation history
Ilke Ileri0385f33 -
Enhance request logging for better Vapi debugging
Ilke Ileri260002e -
Add keyword-based sales filter to redirect off-topic questions
Ilke Ileri7242003 -
Fix: revert system prompt, add full OpenAI-compatible response format
Ilke Ileria1e3c35 -
Add system prompt guard to enforce sales-only responses
Ilke Ileria73c020 -
Fix CORS issues and add request logging for Vapi integration
Ilke Ilerie48c956 -
Improve response formatting and generation parameters
Ilke Ilerie683a4c -
Add HF_TOKEN support for gated model access
Ilke Ileri25839d0 -
Add Vapi Gemma API application
Ilke Ileri692ef6b -
initial commit
Verified signature3ea09f6