Skip to main content
Version: v3.0.0

Kokoro-FastAPI

Run Kokoro-FastAPI locally and connect it to OpenReader using the Custom OpenAI-Like provider.

warning

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

Run Kokoro​

CPU:

docker run --name kokoro-tts \
  --restart unless-stopped \
  -d \
  -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 (NVIDIA):

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

Connect to OpenReader​

Recommended (auth + admin): Settings → Admin → Shared providers

  1. Add a shared provider with type custom-openai.
  2. Set base URL to your Kokoro endpoint (e.g. http://kokoro-tts:8880/v1).
  3. Leave API key blank unless required by your deployment.
  4. Set default model to Kokoro.

Legacy bootstrap seed (optional, first boot only):

API_BASE=http://kokoro-tts:8880/v1

Use kokoro-tts if that's the container name, or host.docker.internal if not.

Or in-app via Settings → TTS Provider:

  1. Set provider to Custom OpenAI-Like.
  2. Set API_BASE to your Kokoro endpoint (e.g. http://kokoro-tts:8880/v1).
  3. Leave API_KEY blank unless your deployment requires one.
  4. Choose model Kokoro.

See TTS Providers for admin-shared vs per-user behavior.

References​