Cline Setup with Claude and GPT via a Custom API Key

Cline is an open-source AI coding agent that runs inside VS Code. It reads your files, proposes edits, runs terminal commands with your approval and works through a task step by step. Cline does not include a model of its own: you bring an API provider and a key, and every step of a task is a request billed by that provider.

That makes the choice of provider the main factor in what Cline costs you. APIVAI gives you one key that works with Claude and GPT models, billed per token below official list prices, from a balance you top up when you need to. One key works in both the OpenAI format and the Anthropic format, so it fits either of Cline's provider options.

Cline offers two provider types that work with APIVAI:

Pick OpenAI Compatible if you want to switch between Claude and GPT or use a model ID that is newer than Cline's built-in list. Pick Anthropic if you only use Claude and prefer Cline's native Claude integration. You can switch between them at any time; the same key works in both.

1. Sign up at apivai.com with your email.
2. Top up your balance from $10 by card, crypto, Alipay or WeChat Pay. There is no free trial; you pay only for what you use.
3. Open the Dashboard, create a key and copy it. You can give the key its own budget limit so Cline can never spend more than you planned.

1. In VS Code, open the Cline panel and click the settings (gear) icon.
2. In API Provider, choose OpenAI Compatible.
3. Fill in the fields as in the table below.
4. Click Verify if your version shows the button, then save and start a task.

The Base URL must end in /v1 exactly once: Cline appends /chat/completions to it. The model configuration also has Context Window, Image Support and Computer Use; leave them at their defaults unless you know you need to change them.

The price fields only affect the cost estimate Cline shows for a task. Filling them in with APIVAI's prices makes that estimate match what you are actually billed; your real usage is always on the Dashboard.

To see which model IDs your key can use, run:

The model list changes over time, so copy IDs from this response or from the pricing page rather than from old tutorials.

1. In API Provider, choose Anthropic.
2. Paste your APIVAI key into Anthropic API Key.
3. Tick Use custom base URL and enter https://api.apivai.com.
4. Choose a Claude model from the Model dropdown.

Do not add /v1 here. Cline's Anthropic client adds /v1/messages to the base URL itself, so https://api.apivai.com/v1 would turn into /v1/v1/messages and fail with 404. If the Claude model you want is not in Cline's list yet, use the OpenAI Compatible provider and type its ID. More about this format is on the Claude API proxy page.

Cline sends many requests per task, so the model's price per token matters more than in a chat app. Prices per 1M input / output tokens as of 2026-10-07:

A practical setup: use Claude Sonnet 4.6 or Claude Sonnet 5.5 by default, switch to Claude Opus 5.5 when a task keeps failing, and use Claude Haiku 4.5 for renames, comments and small fixes. GPT models only work through the OpenAI Compatible provider.

Each step of a Cline task resends the conversation and the files Cline has read, so input tokens add up quickly. Two realistic examples with Claude Sonnet 4.6:

• One medium task, such as adding a feature across a few files: about 20 requests of 30,000 input and 1,500 output tokens each. Cost: $0.781 with APIVAI, $2.25 at official prices.
• A working month of three such tasks a day for 22 days (1,320 requests): $51.52 with APIVAI, $149 at official prices.

The same month on Claude Haiku 4.5 would cost $17.31. These estimates ignore prompt caching and assume thinking tokens are included in the output figure; thinking tokens are billed as output. The Dashboard lists every request with its tokens and cost, so after a day you can replace these assumptions with your own numbers. For a wider comparison, see Claude API pricing in 2026.

• 401 Unauthorized / invalid API key: the key was copied incompletely, deleted, or has reached its budget limit. Check it on the Dashboard. With the Anthropic provider, also make sure Use custom base URL is ticked; otherwise Cline sends your key to Anthropic directly.
• 404 Not Found: almost always the /v1 part. OpenAI Compatible needs https://api.apivai.com/v1 (one /v1, no trailing /chat/completions). Anthropic needs https://api.apivai.com with no /v1.
• Model not found: the Model field must contain an exact ID such as claude-sonnet-4-6, not a display name. Check the ID against GET /v1/models.
• Empty or cut-off replies: thinking is on by default and uses output tokens. Raise Max Output Tokens to 8192 (at least 4096).
• 429 Too Many Requests: the default limit is 60 requests per minute per key. Long agent runs can hit it; wait a moment or ask support to raise the limit.

The general format rules are on the OpenAI-compatible API page and in the docs.

Yes. One APIVAI key works for every model. Use the OpenAI Compatible provider and change the Model field to switch between claude-sonnet-4-6 and gpt-5.5.

OpenAI Compatible covers Claude and GPT with one setup and lets you type any model ID. The Anthropic provider is for Claude only and uses Claude's native format; enter the base URL without /v1.

No. The content of requests and responses is not logged. Only usage data such as model, token counts and cost is kept for billing.

No subscription and no free credits. You top up from $10 and pay only for the tokens you use.

Yes. The same key works in Roo Code, Cursor and Claude Code. Creating a separate key per tool makes the Dashboard easier to read.

Create an account, top up from $10 and paste your key into Cline.