Commit History

  1. fix: improve graph generation reliability (strip thinking blocks, add retry, robust DOT extraction)

    ego
    1fe2fca
  2. fix: disable thinking mode for graph generation to fix corrupted DOT output

    ego
    4822069
  3. fix: switch to langchain-chroma + pin chromadb==0.5.23 to fix KeyError ephemeral

    ego
    63d3704
  4. fix: use chromadb.EphemeralClient() to resolve KeyError ephemeral

    ego
    315fadc
  5. Revert LLM back to nvidia/nemotron-3-nano-30b-a3b, update UI description

    ego
    e4170ab
  6. Switch LLM to google/gemma-4-31b-it, update all model descriptions in UI

    ego
    a9d5e01
  7. Fix TTS synthesize() parameter name: use voice_name to match gRPC protocol

    ego
    52e8dd2
  8. Fix TTS: use correct TTSService class, add fallback when TTS returns None

    ego
    19edd1b
  9. Simplify podcast to single-voice narration: remove dual-speaker format from prompt and comments

    ego
    468cac0
  10. Switch TTS from Google Gemini to NVIDIA Riva TTS (magpie-tts-multilingual hosted API)

    ego
    08a9522

  1. Switch embedding model from Google to NVIDIA NIM (llama-nemotron-embed-1b-v2)

    ego
    cf10bac

  1. final

    ego
    4681215
  2. update name

    ego
    b72dacf

  1. update home

    ego
    737b76d
  2. update prompt

    ego
    6882231
  3. without comment

    ego
    7bafc8f

  1. 1.0

    ego
    ba7bcd3
  2. HF-Deployment: Initial commit

    ego
    ea8f8db