Skip to main content
Version: v2.0.0

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.

Provider​

  • Provider: Custom OpenAI-Like
  • Typical model: Kokoro
  • API_BASE: required (typically your Kokoro URL ending with /v1)
  • API_KEY: set only if your deployment requires one

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

Run Kokoro (GPU)​

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

OpenReader setup​

  1. Start Kokoro using either the CPU or GPU image.
  2. In OpenReader Settings, choose provider Custom OpenAI-Like.
  3. Set API_BASE to your Kokoro endpoint (for Docker Compose, commonly http://kokoro-tts:8880/v1).
  4. Set API_KEY only if your deployment requires one.
  5. Choose model Kokoro.

Notes​

Runtime guidance

GPU mode requires NVIDIA Docker support and is best on NVIDIA hardware. CPU mode is a good default on Apple Silicon and modern x86 CPUs.

References​