Integrations
Kilo Code
Run Kilo Code's agent on any model in the MiniRouter catalog.
Kilo Code is an open-source coding agent for VS Code, JetBrains and the terminal. Add MiniRouter as a custom provider to run it on any tool-capable model in the catalog, with one balance and per-key spend limits.
1. Install Kilo Code
code --install-extension kilocode.Kilo-Codenpm install -g @kilocode/cli2. Create and check a key
Create a key, export it as MINIROUTER_KEY, then send one request. A JSON reply means the key and balance work.
curl https://api.minirouter.sh/v1/chat/completions \
-H "Authorization: Bearer $MINIROUTER_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"zai/glm-5.3-flash","messages":[{"role":"user","content":"ping"}]}'3. Add the provider
Open Kilo Code ⚙ Settings → Providers → Custom provider and fill in:
- Provider ID
minirouter- Provider API
- OpenAI Compatible
- Base URL
https://api.minirouter.sh/v1Never append /chat/completions.- API key
mr-live-… (your key)- Models
zai/glm-5.3-flashPick from the list Kilo loads from /v1/models, or type the ID.
4. Send a prompt
Pick a MiniRouter model, send one prompt, then open Activity. The request lists its model, tokens and exact cost.
Pick models
Choose from the loaded list, or type any ID.
- zai/
glm-5. 3-flash - Everyday coding. See GLM 5.3 Flash.
- openai/
gpt-5. 6-sol - Hard plans and reviews. See GPT 5.6 Sol.
- minirouter/
auto: code - Picks a model with reliable tool calling. See Auto router.
- @preset/
your-slug - Your saved models and instructions. See Presets.
Config files
The CLI, VS Code and JetBrains read the same files. The project file wins.
- Global
~/.config/kilo/kilo.jsonc- Project
.kilo/kilo.jsoncorkilo.jsoncin the repo root.
Set limit.context and limit.output on each model, or Kilo never compacts. Copy both from the model's page in Models.
Run headless
kilo run runs one task without the TUI. The model is the provider ID, a slash, then the model ID.
kilo run --auto --model minirouter/zai/glm-5.3-flash "Run tests and fix failures"- --auto
- Approves actions without asking. Without it, the run rejects them.
- --model
- Provider ID and model ID, for example
minirouter/minirouter/auto:code.
Recommended modelsBrowse every model in the catalog.
zai/glm-5.3-flash- Best current balance of coding ability, agentic performance, context and cost.
openai/gpt-5.6-sol- Higher-intelligence option when task quality matters more than spend.
spacexai/grok-4.6- Strong agentic alternative with a 500K-token window.
google/gemini-3.8-flash- Multimodal coding alternative with a 1M-token window.
Troubleshooting
404, HTML, or a parse error on every request
Almost always the base URL. Use https://api.minirouter.sh/v1 — the bare host without /v1 also works, but never append /chat/completions to the base URL field; the client adds that path itself.
Error: 400 invalid_request.
401 on every request
Wrong or rotated key. Keys start with mr-live-; after a rotation the old key keeps working for 24 hours, then dies.
Error: 401 invalid_api_key.
402 before any tokens arrive
Balance below the estimated cost of the request. Top up (from $0.50 with direct Solana payments) — and watch the x-minirouter-balance-usd header, the low-balance signal every response carries.
Error: 402 insufficient_credits.
Response stops mid-sentence with a credits message
Balance hit $0 mid-stream. The stream ends with a terminal error naming the exact charge for tokens delivered — never a silent close.
Error: 402 insufficient_credits.
Every error code, with its fix: Errors.