Running MCP 41 Serverless TextGen Hub ♨ 41 Now supports inference providers and multimodal. No GPU req.
