MegaBrain University · guide

OpenCode Custom Provider: Use Claude, GPT-6, Kimi and GLM via an OpenAI-Compatible API

Updated November 11, 2026 · 10 min setup

What is OpenCode

OpenCode is an open-source AI coding agent. It runs in the terminal (TUI), as a desktop app and in IDEs, edits files, runs commands, uses subagents and MCP servers, and has a server mode that other apps (OpenWork, for one) build on.

OpenCode works with many providers, and any OpenAI-compatible API can be added as a custom provider. With MegaBrain you get every major coding model behind one key: Claude for hard refactors, GPT-6 as a second opinion, Kimi, GLM and DeepSeek for cheap high-volume edits.

Setup, step by step

1. Step 1. Get a MegaBrain API key

   Sign up with GitHub, Google or email, add funds on the Balance page and create a key on the API key page. The key works for every text, image and video model, and each request is paid from your balance — no subscription.

   Put the key in an environment variable so it never lands in a config file you share:

   # ~/.zshrc or ~/.bashrc
   export MEGABRAIN_API_KEY="your-megabrain-key"

   Copy

   curl https://getmegabrain.com/api/gateway/v1/chat/completions \
     -H "Authorization: Bearer $MEGABRAIN_API_KEY" \
     -H "Content-Type: application/json" \
     -d '{"model": "anthropic/claude-sonnet-5.5", "messages": [{"role": "user", "content": "Say OK"}]}'

   Copy

   Check the key before you touch the agent. A JSON reply with choices means the key, the balance and the model all work:

2. Step 2. Install OpenCode

   Any one of:

   curl -fsSL https://opencode.ai/v2/install | bash
   # or
   npm i -g @opencode/cli@latest
   # or (macOS)
   brew install opencode

   Copy

3. Step 3. Add MegaBrain to opencode.json

   Put this in ~/.config/opencode/opencode.json (all projects) or opencode.json in a project root. Only models listed under models can be selected:

   {
     "$schema": "https://opencode.ai/config.json",
     "model": "megabrain/anthropic/claude-sonnet-5.5",
     "provider": {
       "megabrain": {
         "npm": "@ai-sdk/openai-compatible",
         "name": "MegaBrain",
         "options": {
           "baseURL": "https://getmegabrain.com/api/gateway/v1",
           "apiKey": "{env:MEGABRAIN_API_KEY}"
         },
         "models": {
           "anthropic/claude-sonnet-5.5": {
             "name": "Claude Sonnet 5.5",
             "limit": { "context": 1000000, "output": 32000 }
           },
           "anthropic/claude-opus-5.5": {
             "name": "Claude Opus 5.5",
             "limit": { "context": 1000000, "output": 32000 }
           },
           "openai/gpt-6.1-sol": {
             "name": "GPT-6.1 Sol",
             "limit": { "context": 1050000, "output": 32000 }
           },
           "moonshotai/kimi-k3.1-pro": {
             "name": "Kimi K3.1 Pro",
             "limit": { "context": 1048576, "output": 32000 }
           },
           "z-ai/glm-5.3": {
             "name": "GLM 5.3",
             "limit": { "context": 1048576, "output": 32000 }
           }
         }
       }
     }
   }

   Copy

   This is the classic (V1) config shape. OpenCode V2 documents a new shape (providers, package, settings.baseURL) but still reads this one and converts it in memory, and OpenWork and the Kilo CLI read the same file — so one config works everywhere.

4. Step 4. Start OpenCode and pick the model

   Run opencode in your project. The default model comes from "model"; switch with /models (MegaBrain models are grouped under the provider name). Ask for something that touches files — “add a test for utils/date.ts” — to confirm tool calls work.

Which model to choose

Always set limit.context and limit.output: OpenCode uses them to decide when to compact the conversation. Claude Sonnet 5.5 is the default for coding; use Opus 5.5 for large refactors and Kimi, GLM or DeepSeek Flash for cheap repetitive edits.

Every model with its price and context window is in the catalog. They all work with the same key.

What you can do

Business use cases

Replace every use of moment.js in src/ with date-fns. Keep behaviour identical, update the tests, run npm test and fix failures. Commit nothing; list the changed files at the end.

Troubleshooting

FAQ