Cline + Kunavo — OpenAI Compatible provider with a correct cost readout

API Provider OpenAI Compatible Base URL https://api.kunavo.com/v1 API Key sk-kn-... Model ID claude-sonnet-5 Model Configuration Context Window 200000 Max Output Tokens 64000 Image Support on Input Price (from the table below) Output Price (from the table below)

Step by step

Create a key at /app/keys and copy it — it is shown once.

Open the Cline settings panel (the ⚙︎ icon) and choose OpenAI Compatible as the API Provider.

Enter the Base URL, the API Key and the Model ID.

Expand Model Configuration and set the context window, max output tokens, image support and both prices.

Start a task. Plan/Act can hold different models, so pair a planning model with a cheaper acting one rather than paying the planning rate for every edit.

Checked against Cline's OpenAI Compatible provider page on September 6, 2026. Third-party settings move; if a field name here no longer matches what you see, that page is the authority, not this one.

This is the short version. The full walkthrough — model choice, what a real session costs, and the failure modes — is in the Cline + Claude walkthrough.

Verify before you debug the client

One request settles whether a failure is the endpoint, the key, or the configuration file. If this returns JSON, the same base URL and key work in Cline.

# Settles whether a failure is the endpoint, the key, or the client.
curl -sS https://api.kunavo.com/v1/models \
  -H "Authorization: Bearer sk-kn-..."

Which model id to put in the field

Every text model is reachable as a model id — the live list is GET /v1/models, and the catalog with prices is on the models page. Rates are USD per 1M tokens, input / output.

Model id	Kunavo in / out	Where it fits in Cline
`claude-sonnet-5`	$2.00 / $10.00	Act mode — the default working model
`claude-opus-5`	$2.00 / $10.00	Plan mode, where the plan is what the rest of the task pays for
`claude-haiku-4-5`	$0.40 / $2.00	cheap reads and repeated file inspection
`gpt-5-4`	$1.00 / $6.00	a different family behind the same key, for a second opinion

Billing is per token from a prepaid balance with no monthly fee — see billing. On repeated context — which is most of what an editor or a chat client sends — prompt caching moves the bill more than the model choice does.

FAQ

How do I use Cline with a custom OpenAI-compatible API?

In the Cline settings panel select "OpenAI Compatible" as the API Provider, then enter the endpoint's Base URL, your API Key and the Model ID you want. The base URL is the /v1 root, because Cline appends only the route to it. Cline then speaks the OpenAI wire format to that endpoint regardless of which vendor actually serves the model.

Can Cline use Claude models without an Anthropic API key?

Yes, when the models are served over an OpenAI-compatible endpoint. Cline sends the Model ID through unchanged, so a Claude id is resolved by the endpoint rather than by the extension, and the only credential Cline holds is the endpoint's. That is also how one key can cover Claude, GPT and Gemini ids in the same provider entry.

Why is Cline's cost estimate wrong?

Because Cline computes it from the Input Price and Output Price fields in the model configuration block, which are entered by hand and are not fetched from the endpoint. If they were left at defaults or copied from a different provider, every cost readout inherits that error. Fill them with the rates the endpoint actually charges for the model id you selected.

Should Cline's base URL end in /v1?

Yes for an OpenAI-compatible endpoint. Cline appends the route — /chat/completions — to whatever base URL you give it, so https://api.example.com/v1 resolves and https://api.example.com sends the request to a path that does not exist. A 404 points at the base URL; a 401 points at the key.