config.yaml

huggingface:
  # Qwen/Qwen3-235B-A22B-Thinking-2507-FP8
  # Qwen/Qwen3-235B-A22B-Thinking-2507
  # Qwen/Qwen3-235B-A22B-Instruct-2507-FP8
  # Qwen/Qwen3-235B-A22B-Instruct-2507
  # Qwen/Qwen3-30B-A3B-Thinking-2507-FP8
  # Qwen/Qwen3-30B-A3B-Thinking-2507
  # Qwen/Qwen3-30B-A3B-Instruct-2507-FP8
  # Qwen/Qwen3-30B-A3B-Instruct-2507
  # Qwen/Qwen3-4B-Thinking-2507-FP8
  # Qwen/Qwen3-4B-Thinking-2507
  # Qwen/Qwen3-4B-Instruct-2507-FP8
  # Qwen/Qwen3-4B-Instruct-2507
  # Qwen/Qwen3-32B

  chat_model: Qwen/Qwen3-30B-A3B-Instruct-2507
  embedding_model: Qwen/Qwen3-Embedding-8B
  # Qwen/Qwen3-Reranker-4B
  reranker_model: Qwen/Qwen3-Reranker-8B

chunking:
  chunk_size: 1000
  chunk_overlap: 200
  separators: ["\n\n", "\n", " "]


apis_models:
  num: 2
  ###   silicon_flow    ###
  silicon_flow:
    qwen:
      chat3_30b: Qwen/Qwen3-30B-A3B-Instruct-2507
      qwen_14B: Qwen/Qwen1.5-14B-Chat
      embed: Qwen/Qwen3-Embedding-8B
      rerank: Qwen/Qwen3-Reranker-8B

    fallback:
      openai:
        api_chat:


  ###      groq     ###
  groq:
    qwen:
      chat3_32b: Qwen/Qwen3-32B ## 32b is working properly
      qwen_14B: Qwen/Qwen3-14B-Chat
      embed: Qwen/Qwen3-Embedding-8B
      rerank: Qwen/Qwen3-Reranker-8B

    openai:
      gpt_oss: openai/gpt-oss-120b
      
  ###      deepinfra     ###
  deepinfra:
    qwen:
      rerank: Qwen/Qwen3-Reranker-8B
      embed: Qwen/Qwen3-Embedding-8B
      chat: Qwen/Qwen3-30B-A3B-Instruct-2507


rag:
  weavaite_collection: Aya  # updates (8/09) Books 
  old_doc: Old_Documents


groq:

# Audio Configuration
audio:
  model_live: gemini-2.0-flash-live-001
  tts_model: gemini-2.5-flash-preview-tts
  stt_model: gemini-2.0-flash
  VAD_RATE: 16000
  output_sample_rate: 24000
  vad_mode: 3
  VAD_FRAME_MS: 20
  padding_ms: 300
  vad_ratio: 0.9
  default_voice: "Kore"
  available_voices:
    - "Kore"
    - "Puck"
    - "Zephyr"
    - "Leda"
    - "Fenrir"
    - "Charon"
    - "Orus"
    - "Aoede"
    - "Callirrhoe"