Or jump from this list on a smaller screen.

This is a page from the Audivra documentation. Developer exportsindex: http://127.0.0.1:3000/docs/llms.txt, full corpus: http://127.0.0.1:3000/docs/llms-full.txt, OpenAPI: http://127.0.0.1:3000/docs/openapi.json, Postman: http://127.0.0.1:3000/docs/postman.json. Press ⌘K to search docs.

Processing nodes

Short prompts route to local Ollama (LLM), Kokoro/Piper (TTS), and MuseTalk/Wav2Lip (lip-sync) before offloading to GPU cluster or external APIs. Every execution is logged in PostgreSQL with atomic credit deduction.

local_worker

Ollama LLM, Kokoro/Piper TTS, MuseTalk on your machine or RTX GPU.

When used: Prompt under 500 chars, local worker healthy, user prefers fast path.

Routing cost: 1 credit(s)

gpu_cluster

Celery workers or Modal T4/A10G serverless GPUs.

When used: Long scripts, lip-sync video jobs, or local VRAM saturated.

Routing cost: 3 credit(s)

external_api

Optional third-party fallback when EXTERNAL_API_URL is configured.

When used: forceCloud=true or failover from local worker.

Routing cost: 3 credit(s)

Open-source stack

  • Fast LLM: Ollama — qwen2.5:3b / llama3.2:3b (~2.5 GB VRAM/RAM)
  • TTS: Kokoro-82M or Piper TTS (~500 MB RAM / CPU-friendly)
  • Lip-sync video: MuseTalk or Wav2Lip (~4 GB VRAM)

Setup guide: Local worker · API: POST /api/v1/generate