Cline setup: use Claude Sonnet 5.5 with an OpenAI-compatible API key
Cline talks to any OpenAI-compatible endpoint. Point it at MiniStaff API and Claude Sonnet 5.5 runs on a prepaid key at $1.00 / $5.00 per 1M tokens — exactly half the official price — with no subscription and no weekly meter.
Before you start
- A MiniStaff account with a topped-up wallet — sign up at app.ministaff.pro ($10 minimum; in India, top up by UPI from ₹950).
- An API key from the dashboard — it starts with
msst_. - Cline installed in VS Code.
The three settings
Open Cline, go to Settings → API Configuration, and set:
| Field | Value |
|---|---|
| API Provider | OpenAI Compatible |
| Base URL | https://api.ministaff.pro/v1 |
| API Key | Your msst_... key |
| Model ID | claude-sonnet-5-5 |
That is the whole configuration. Cline sends its requests to /v1/chat/completions with streaming — exactly the endpoint MiniStaff serves. If Cline shows optional model-information fields (context window, max tokens), the defaults work; leave them as they are.
Check it works
- Start a new Cline task and send a small message, for example: “Reply with exactly: OK” — you should get a short streamed reply.
- Open your MiniStaff dashboard — the call appears there with its model, token counts and cost, so you can watch the meter yourself.
Prefer to test outside the editor first? The same request Cline makes, from a terminal:
curl https://api.ministaff.pro/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer msst_your_key_here" \
-d '{
"model": "claude-sonnet-5-5",
"messages": [{"role": "user", "content": "Reply with exactly: OK"}]
}'
Use any of the 19 models
Sonnet is the starting point, not the limit. Change the Model ID field to any ID from the model list — cheaper workhorses for bulk edits, frontier models for hard problems:
| Model ID | Input / 1M | Output / 1M | Good for |
|---|---|---|---|
deepseek-v4.1-flash | $0.075 | $0.30 | Cheap, fast everyday edits |
gpt-6-luna | $0.05 | $0.25 | Lightweight tasks |
claude-sonnet-5-5 | $1.00 | $5.00 | Serious coding work |
claude-opus-5-5 | $2.00 | $10.00 | The hardest problems |
The live list is always at GET /v1/models, with the full rate card in the docs. Two billing details worth knowing: cached input is billed at 1/10 of the normal input rate, and the rate limit is 60 requests per minute per account.
Troubleshooting
| Error | What it means | Fix |
|---|---|---|
| 401 | Key missing, mistyped or revoked | Re-paste the key from the dashboard, or create a new one |
402 insufficient_quota | Wallet balance is $0 | Top up — the same key resumes working |
| 429 | Rate limit (60 requests/minute) | Wait a few seconds and retry |
| Model not available | Model ID mistyped | Copy the ID exactly, e.g. claude-sonnet-5-5 |
Still stuck? Write to [email protected].