Kokoro-FastAPI/compose.gpu.yaml at master · smittie2000/Kokoro-FastAPI · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
# Docker Compose for Coolify deployment — Qwen3-TTS (GPU)
# Requires NVIDIA GPU + Container Toolkit on the host.
#
# Deploy as a separate Coolify resource:
#   Repository: tts-service repo
#   Compose file: compose.gpu.yaml

services:
  tts-qwen3:
    build:
      context: .
      dockerfile: docker/qwen3/Dockerfile
    restart: unless-stopped
    labels:
      - "app=${APP_NAME:-eloise}"
      - "service=tts-qwen3"
    expose:
      - "8880"
    environment:
      - 'TTS_ENGINE=qwen3'
      - 'USE_GPU=true'
      - 'HF_HOME=/app/models'
      - 'API_KEYS=${API_KEYS:-}'
      - 'DEBUG_API_KEY=${DEBUG_API_KEY:-}'
      - 'ENABLE_WEB_PLAYER=${ENABLE_WEB_PLAYER:-false}'
      - 'ENABLE_DOCS=${ENABLE_DOCS:-false}'
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    volumes:
      - tts_qwen3_models:/app/models
      - tts_qwen3_voices:/app/voices/qwen3
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8880/health"]
      interval: 15s
      timeout: 5s
      retries: 5
      start_period: 180s

volumes:
  tts_qwen3_models:
    driver: local
  tts_qwen3_voices:
    driver: local