Connect

vLLM-Omni API Integration

hsliuustc0106byhsliuustc0106
•
74
•
API Development

Integrates Claude with vLLM-Omni servers for multi-modal text, image, video, and audio generation via OpenAI-compatible endpoints.

The vLLM-Omni API skill empowers Claude to interact seamlessly with vLLM-Omni's multi-modal capabilities using standardized OpenAI-compatible REST endpoints. It provides structured guidance for implementing chat completions, image generation, image editing, and speech synthesis within client applications. By streamlining the communication between Claude and local or remote vLLM-Omni instances, this skill facilitates the development of complex AI tools that handle diverse inputs and outputs including text, vision, and audio.

Key Features

01Support for multi-modal inputs including text, image, and audio
02OpenAI-compatible REST API integration for all modalities
03Direct image generation and advanced image editing capabilities
04Text-to-speech synthesis with custom voice cloning support
05Asynchronous video generation with polling mechanisms
0674 GitHub stars

Use Cases

01Building multi-modal client applications powered by local AI models
02Integrating text-to-speech and voice cloning into AI-driven agents
03Automating complex image and video generation workflows