danvics/openreader
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
openreader/docs-site/docs/integrations/kokoro-fastapi.md
Richard R 6d5fb65a72 feat(docs): add configurable rate limiting and external docs 
Add environment variables for fine-grained control over TTS rate limiting
and Better Auth behavior. Move documentation to external Docusaurus site
with automated deployment workflows.

- TTS rate limiting can now be enabled/disabled via TTS_ENABLE_RATE_LIMIT
- Customizable daily limits for anonymous/authenticated users and IP backstops
- Better Auth rate limiting can be disabled via DISABLE_AUTH_RATE_LIMIT
- Rename library import env vars to IMPORT_LIBRARY_DIRS/DIR
- Add docs-site with Docusaurus and GitHub Actions workflows
- Update README to reference external documentation
2026-02-10 15:11:37 -07:00

1.3 KiB
Raw Blame History

title
Kokoro-FastAPI

You can run the Kokoro TTS API server directly with Docker.

:::warning For Kokoro issues and support, use the upstream repository: remsky/Kokoro-FastAPI. :::

CPU image

docker run -d \
  --name kokoro-tts \
  --restart unless-stopped \
  -p 8880:8880 \
  -e ONNX_NUM_THREADS=8 \
  -e ONNX_INTER_OP_THREADS=4 \
  -e ONNX_EXECUTION_MODE=parallel \
  -e ONNX_OPTIMIZATION_LEVEL=all \
  -e ONNX_MEMORY_PATTERN=true \
  -e ONNX_ARENA_EXTEND_STRATEGY=kNextPowerOfTwo \
  -e API_LOG_LEVEL=DEBUG \
  ghcr.io/remsky/kokoro-fastapi-cpu:v0.2.4

GPU image

docker run -d \
  --name kokoro-tts \
  --gpus all \
  --user 1001:1001 \
  --restart unless-stopped \
  -p 8880:8880 \
  -e USE_GPU=true \
  -e PYTHONUNBUFFERED=1 \
  -e API_LOG_LEVEL=DEBUG \
  ghcr.io/remsky/kokoro-fastapi-gpu:v0.2.4

OpenReader integration notes

  • In OpenReader settings, choose provider Custom OpenAI-Like and model Kokoro.
  • Set OpenReader API_BASE to your Kokoro endpoint (for Docker Compose, commonly http://kokoro-tts:8880/v1).
  • API_BASE is needed here because Kokoro is used via the custom provider path, not a built-in provider endpoint.
  • GPU mode requires NVIDIA Docker support and is best on NVIDIA hardware.
  • CPU mode works best on Apple Silicon or modern x86 CPUs.