{ "$schema": "https://opencode.ai/config.json", "permission": "allow", "mcp": { "gitea": { "type": "local", "command": ["go", "run", "gitea.com/gitea/gitea-mcp@latest", "-t", "stdio"], "environment": { "GITEA_HOST": "https://git.cleverthis.com", "GITEA_ACCESS_TOKEN": "{env:FORGEJO_PAT}" }, "enabled": true }, "context7": { "type": "local", "command": ["npx", "-y", "@upstash/context7-mcp"], "enabled": false }, "sequential-thinking": { "type": "local", "command": ["npx", "@modelcontextprotocol/server-sequential-thinking"], "enabled": false }, "docker": { "type": "local", "command": ["npx", "mcp-server-docker"], "enabled": false }, "sqlite": { "type": "local", "command": ["uvx", "mcp-server-sqlite"], "enabled": false }, "postgres": { "type": "local", "command": ["npx", "-y", "@modelcontextprotocol/server-postgres"], "environment": { "POSTGRES_CONNECTION_STRING": "{env:POSTGRES_CONNECTION_STRING}" }, "enabled": false }, "playwright": { "type": "local", "command": [ "npx", "@playwright/mcp@latest", "--browser=firefox", "--caps=storage,testing,vision", "--headless" ], "timeout": 300000, "enabled": false } }, "provider": { "llama-local": { "name": "Llama.cpp (10.233.1.16)", "npm": "@ai-sdk/openai-compatible", "options": { "baseURL": "http://10.233.1.16:8080/v1" }, "models": { "google-gemma-4-31b-q5km-text": { "name": "Gemma 4 31B text", "limit": { // adjust this according to llama-server parameter "context": 262144, "output": 65536 }, "options": { "temperature": 1.5 // according to reddit, this temp works better than the recommended 1.0 for coding }, "modalities": { "input": ["text"], "output": ["text"] } }, "google-gemma-4-26b-a4b-q80-vision": { "name": "Gemma 4 26B A4B Vision", "limit": { // adjust this according to llama-server parameter "context": 262144, "output": 65536 }, "options": { "temperature": 1.0 }, "modalities": { "input": ["text", "image"], "output": ["text"] } } } }, "bifrost": { "npm": "@ai-sdk/openai-compatible", "options": { "baseURL": "http://10.233.1.16:1234/v1" }, "models": { "ct/qwen3.5-9b": { "name": "Qwen 3.5 9B (Tier 1)", "limit": { "context": 233216, "output": 65536 }, "options": { "temperature": 0.6 }, "modalities": { "input": ["text", "image"], "output": ["text"] } }, "ct/qwen3.6-35b-a3b": { "name": "Qwen 3.6 35B A3B (Tier 2)", "limit": { "context": 262144, "output": 65536 }, "options": { "temperature": 0.6 }, "modalities": { "input": ["text", "image"], "output": ["text"] } }, "ct/minimax-m2.7": { "name": "MiniMax M2.7 (Tier 3)", "limit": { "context": 113664, "output": 65536 }, "options": { "temperature": 1.0 }, "modalities": { "input": ["text"], "output": ["text"] } }, "ct/kimi-k2.6": { "name": "Kimi K2.6 (Tier 4)", "limit": { "context": 262144, "output": 65536 }, "options": { "temperature": 1.0 }, "modalities": { "input": ["text", "image"], "output": ["text"] } }, "ct/qwen3-coder-480b-a35b-inst": { "name": "Qwen3 Coder 480B A35B (Tier 5)", "limit": { "context": 861952, "output": 65536 }, "options": { "temperature": 0.7 }, "modalities": { "input": ["text"], "output": ["text"] } } } } } }