API429 / Cline

API429 in Cline: OpenAI Compatible settings and a bounded first check

Select OpenAI Compatible in Cline and enter the API429 base URL, your client key and an exact model ID. Begin with a small text check. The provider setting describes a connection format, while tools, file actions and complete coding workflows require additional validation.

Configuration is based on the application’s public documentation and the API429 contract. Native client execution, installation imports and agent workflows have not been verified here. Start with a short non-sensitive request and inspect the response in your installed version.

Connection settings

Protocol
OpenAI Chat Completions
Connection mode
OpenAI Compatible
Base URL
https://gateway.api429.com/v1
Exact model ID to check
gpt-6.1-sol

Check access for your key with GET /v1/models. A public model card does not guarantee availability in your application. Store the key in the application’s secure credential store or the current session environment. Keep secret values out of command arguments and shared project files.

Prepare the model and a disposable workspace

Use an installed Cline client, an API429 key and a model available to that key. The example is gpt-6.1-sol. Check GET /v1/models before choosing it, because a public card is not a key-specific access guarantee. Start with an empty test directory or a small nonsecret text file rather than granting access to a production repository for an endpoint check.

Open settings with the Cline gear control and choose OpenAI Compatible. The model family name alone does not select a transport: GPT or Gemini branding is not proof of a particular request method. Native Cline execution was not performed for this guide. A first request may be paid, and a successful text response would not establish a complete agent workflow.

Enter the connection fields carefully

Use https://gateway.api429.com/v1 as Base URL, save your API429 key in the API Key field and enter gpt-6.1-sol as Model ID. Use Custom Model ID where that is the field offered by your version. Do not paste the full /v1/chat/completions path into the base field or append another /v1. The example is a field reference rather than an importable application configuration.

Keep credentials out of shared workspace files, screenshots and diagnostic messages. If your Cline version maintains separate Plan and Act provider settings, inspect both before starting a task. Saving one mode does not establish the other mode uses the same selection. Follow the actual interface labels in your installed build instead of copying an unrelated settings.json from another client.

Do not invent context and output limits

Model Configuration exposes fields such as Context Window Size and Max Output Tokens. These settings help the client manage context and output, but they should use confirmed limits for the exact API429 ID. The reviewed public catalog did not establish effective context and output maxima for this connection, so this guide does not fill those fields with manufacturer marketing numbers.

Obtain current limits for the chosen connection before submitting a large project. An unset output field does not mean unlimited output. Keep the initial question small and use a bounded limit where the client and request method support it. Image support, MCP, computer use and file operations each require a separate check; a checkbox cannot create a model capability.

Check text before enabling agent actions

Ask a short nonconfidential text question first and inspect the response. For 401, check the protected key field and provider selection. For a missing model, compare the literal ID with the authenticated catalog. For a limit error, reduce the supplied context and confirm the effective limits before retrying. Use redacted error details and avoid repeated large requests.

The public catalog marks gpt-6.1-sol with function_calling, but that does not prove the complete Cline tool cycle. The API429 client reference associates GPT 6.1 Sol tooling with Responses API. Selecting OpenAI Compatible does not establish that your client sends that method and format. If it sends Chat Completions tools, verify support before permitting file edits or shell commands.

Measure usage and review a Roo Code migration

Cline estimates depend on the prices and accounting configured in the client. Actual API429 charges depend on requests and token usage, including growing context and repeated actions. Compare actual usage in API429 instead of treating a task estimate as a fixed charge. Cline application terms and any separate software costs remain distinct from API billing.

Roo Code published an official sunset notice: its Extension, Cloud and Router ended on May 15, 2026. Treat that dated shutdown as a migration reason rather than following an old installation guide as current advice. Cline is a separate client to evaluate with the settings and checks above. Do not copy old credentials, assume configuration parity or infer native tool compatibility from a basic OpenAI-compatible text request.

Settings and examples

Cline connection field reference

API Provider: OpenAI Compatible
Base URL: https://gateway.api429.com/v1
API Key: YOUR_API429_KEY
Model ID: gpt-6.1-sol

Usage and next steps

Hypothetical hourly workload: 30 requests to gpt-6.1-sol, each with 5,000 ordinary input and 512 output tokens, costs approximately $0.14 at current API429 rates. Cache, tools, retries, application subscriptions and other fees are excluded. Catalog date: .

Actual usage depends on request count, context length and response size. Check your API429 account for actual charges.

API reference · API429 documentation · Help and diagnostics

Payment and a $10 test

  • Russian bank card. Top up your balance in the account.
  • Invoice for companies. Bank transfer with closing documents.
  • USDT and TON. Top up in crypto.

$10 test credit. Subscribe to our Telegram channel through the bot and get $10 on your balance for the first requests. Get $10 in the bot

Frequently asked questions

Which provider should I select in Cline?

Choose OpenAI Compatible and enter the API429 base URL, protected key and exact model ID. This is separate from native OpenAI or Codex account authorization.

Which context and output numbers should I enter?

Use confirmed effective API429 limits for the exact model. The reviewed public catalog did not establish these maxima, and an unset value does not imply unlimited output.

Does function_calling prove Cline tools work?

No. Check the client method and format as well as the model. The documented GPT 6.1 Sol tool path is Responses API; a basic text response does not validate Cline agent actions.

Should I install Roo Code from an old guide?

Its official sunset notice says Extension, Cloud and Router ended May 15, 2026. Evaluate a current separate client such as Cline and recheck its connection rather than assuming the old setup transfers.

Sources

The application documentation defines its interface and limitations; API429 documentation defines addresses, authentication and client methods. Connection configuration checked: .

Telegram