CosyVoice3-API ghcr.io/hsiang-han/cosyvoice3-api:latest https://ghcr.io bridge bash false https://github.com/hsiang-han/CosyVoice3-API/issues https://github.com/hsiang-han/CosyVoice3-API OpenAI-compatible Text-to-Speech API powered by CosyVoice 3 (Alibaba FunAudioLLM). State-of-the-art Chinese TTS with 18 dialects and 9 languages. NOTE: This model has NO built-in voices. You must register a voice first by uploading a 3-second reference audio via the /v1/voices/register endpoint, or use /v1/audio/speech/clone for one-off cloning. CUDA 12.8 (Blackwell/Ada/Ampere). First start downloads model files (~10GB) from HuggingFace. Supports RTX 50-series GPUs. China users: set HF_ENDPOINT to https://hf-mirror.com for faster downloads. Requires NVIDIA GPU with driver 550+. AI: Productivity: Tools: http://[IP]:[PORT:8080]/docs https://raw.githubusercontent.com/hsiang-han/unraid_templates/main/templates/cosyvoice3-api.xml https://raw.githubusercontent.com/hsiang-han/unraid_templates/main/assets/cosyvoice3-api-icon.png --gpus all --shm-size=2g 8080 /mnt/user/appdata/cosyvoice3-api/models /mnt/user/appdata/cosyvoice3-api/modelscope FunAudioLLM/Fun-CosyVoice3-0.5B-2512 https://huggingface.co true all compute,utility