← Pricing & setup guides

Kilo Code pricing: platform fees, API credits and BYOK explained

Kilo separates the coding platform, model inference and cloud compute. Add those parts before comparing a Pass subscription with your own API provider.

The buying decision

Use the free individual client with your own provider if that fits your workflow. A team seat, an inference subscription and hosted compute are separate budget decisions.

The three bills to distinguish

Kilo's current pricing page lists a free individual platform, a paid Teams platform, separately billed model inference, and separately billed cloud features. Its inference choices include free or local models, bring-your-own-key providers, the Kilo Gateway and Kilo Pass. The platform seat price alone is not the cost of running your agent all month.

Kilo Pass versus another API subscription

Kilo Pass is a monthly credit subscription with bonus-credit terms. A separate provider subscription supplies a different allowance through that provider's API. Compare the total charge, usable credits, model availability, credit purchase fees, expiry and exhaustion behavior. Equal dollar amounts of credit do not guarantee equal completed work: model selection, caching and retries still matter.

Bring Standard Compute to the extension

In Kilo's extension provider settings, use OpenAI Compatible with https://api.stdcmpt.com/v1, your own Standard Compute API key and model ID standardcompute. Verify the exact client surface you intend to use: extension, CLI and hosted cloud agents may expose different settings. A working extension connection does not prove your hosted deployment uses the same provider.

A useful comparison for a small team

List platform seats on one line, inference on another and hosted compute on a third. Run the same task through your present and proposed setup. Compare accepted results and the provider's recorded allowance consumption. Standard Compute team plans assign budgets to paid members; do not assume one member's unused allowance is automatically pooled with another's. Keep the existing provider available while evaluating the replacement.

Run your own comparison

Use the same task, starting revision and acceptance check. Record every retry and failed attempt so the result reflects the work you actually paid for.

Sources and setup references