Spaces:
Runtime error
Runtime error
Commit History
-
fix: improve graph generation reliability (strip thinking blocks, add retry, robust DOT extraction)
ego1fe2fca -
fix: disable thinking mode for graph generation to fix corrupted DOT output
ego4822069 -
fix: switch to langchain-chroma + pin chromadb==0.5.23 to fix KeyError ephemeral
ego63d3704 -
fix: use chromadb.EphemeralClient() to resolve KeyError ephemeral
ego315fadc -
refactor: slim down GRAPH_SYSTEM prompt (~65% token reduction)
ego8632cf1 -
Revert LLM back to nvidia/nemotron-3-nano-30b-a3b, update UI description
egoe4170ab -
Switch LLM to google/gemma-4-31b-it, update all model descriptions in UI
egoa9d5e01 -
Fix TTS synthesize() parameter name: use voice_name to match gRPC protocol
ego52e8dd2 -
Fix TTS: use correct TTSService class, add fallback when TTS returns None
ego19edd1b -
Reduce podcast narration length requirement to 200 words
ego816fcde -
Simplify podcast to single-voice narration: remove dual-speaker format from prompt and comments
ego468cac0 -
Switch TTS from Google Gemini to NVIDIA Riva TTS (magpie-tts-multilingual hosted API)
ego08a9522
-
Switch embedding model from Google to NVIDIA NIM (llama-nemotron-embed-1b-v2)
egocf10bac
-
final
ego4681215 -
update name
egob72dacf
-
update home
ego737b76d -
update prompt
ego6882231 -
Update RAG prompt: bullet points and sequential citations
egoc0a14ba -
Update dependencies: add NVIDIA endpoints
egoe298aef -
Sync HF metadata
ego33710be -
without comment
ego7bafc8f
-
update
ego5acbfd6 -
add ignore
ego6ce2300 -
1.0
egoba7bcd3 -
HF-Deployment: Initial commit
egoea8f8db