One endpoint for every coding agent.

Connect the AI subscriptions you already pay for, then point any OpenAI- or Anthropic-compatible client at a single URL.

Bring the subscriptions you already pay for

Or add any OpenAI- or Anthropic-compatible endpoint, or a model that runs on your own machine.

One URL for every client
OpenAI-compatible Chat Completions and Responses, and Anthropic-compatible Messages. Claude Code, Codex CLI, Cursor, Cline and OpenCode all point at the same place. Endpoints
Failover across accounts
Connect several accounts per provider and set their order. When one reaches a usage limit or fails, the request moves on to the next. Routing
Model groups
Private endpoints with your own model names, each with a failover order across providers and accounts. Groups
Every request on record
Model, account, key, status, tokens, estimated cost and latency for each request. Prompts and replies are never stored. Logs
Teams
Share the accounts you connected with your team, and decide what each member may use. Team
Local models too
Expose Ollama, LM Studio, vLLM or llama.cpp through the CLI's tunnel, without a public address. Local models

Sign in with Google
No password, and nothing to install to begin.
Connect your accounts
Add a subscription, an API endpoint or a local model on the Providers page.
Point your agent at it
Create a key on the Access page and set your agent's base URL, or let the kano-proxy CLI do it.

Every plan includes every feature. A plan sets how many proxy requests you can send.

Free
No request ceiling today
$0
- Proxy requests with no ceiling
- Counted per UTC calendar month
Pro
Room for daily work
$3 USD / month
- 1,000,000 proxy requests per month
- Your allowance resets with your billing period
- Cancel anytime — paid access runs to the end of the period you paid for
Max
Higher limits for heavy use
$10 USD / month
- 5,000,000 proxy requests per month
- 5× the Pro allowance

Payments are processed by Dodo Payments, our Merchant of Record. Terms of Service · Refund Policy