{ "$schema": "https://modelcontextprotocol.io/schema/mcp.json", "name": "venice", "displayName": "Venice AI", "description": "Uncensored, private, crypto-native AI inference (LLM, image, video, TTS, ASR, music, characters) by Venice.ai. Pay-per-call via x402, optional API key, no account required. Community-maintained, provided as-is, no SLA — use at your own risk.", "homepage": "https://venice.ai", "license": "MIT", "version": "0.2.0", "keywords": [ "uncensored", "privacy", "crypto", "x402", "image-generation", "video-generation", "tts", "asr", "music", "characters" ], "transports": [ "stdio" ], "command": { "stdio": { "command": "npx", "args": [ "-y", "@veniceai/mcp-server@0.2.0" ], "env": { "VENICE_API_KEY": "${VENICE_API_KEY}", "VENICE_SIWX_TOKEN": "${VENICE_SIWX_TOKEN}" } } }, "config": { "VENICE_API_KEY": { "description": "Optional Venice API key. When unset, the server passes 402 (x402) challenges through to your MCP host for wallet-based payment.", "required": false }, "VENICE_SIWX_TOKEN": { "description": "Optional base64 SIWE payload for x402 wallet-mode auth. Used when VENICE_API_KEY is unset.", "required": false } } }