> ## Documentation Index
> Fetch the complete documentation index at: https://docs.laozhang.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Cline setup with LaoZhang API

> Set up Cline's OpenAI Compatible provider with LaoZhang API and pick models that call tools reliably in Chat Completions.

Cline calls tools through Chat Completions, where GPT-6 models fail on tool calls, so pick a model from the list below.

| Item | Details |
| - | - |
| API Provider | OpenAI Compatible |
| Base URL | `https://api.laozhang.ai/v1` |
| API Key | Your LaoZhang API key |
| Recommended models | `deepseek-v4-pro`, `gpt-5.6-sol`, `qwen3-coder-flash` |
| Last verified | October 5, 2026 |

## Set it up

<Steps>
  <Step title="Install Cline">
    Search for **Cline** in the VS Code extension marketplace and install it. A Cline icon appears in the activity bar.
  </Step>

  <Step title="Open settings">
    Click the gear icon at the top right of the Cline panel.
  </Step>

  <Step title="Enter the provider details">
    * **API Provider**: `OpenAI Compatible`
    * **Base URL**: `https://api.laozhang.ai/v1`
    * **API Key**: your LaoZhang API key
    * **Model**: the full model ID, such as `deepseek-v4-pro`

    Don't choose the `OpenAI` provider; it always connects to OpenAI's own endpoint.
  </Step>

  <Step title="Adjust model settings (optional)">
    In the model configuration, set Context Window and Max Output Tokens for the model you chose, and check Image Support if it accepts images. If you're unsure, keep the defaults.
  </Step>

  <Step title="Test with a small task">
    Ask Cline to create a file and write one line to it. If it does, chat and tool calls both work.
  </Step>
</Steps>

## Choose a model

Cline needs a model that returns tool calls reliably. These models have been confirmed to call tools in streaming Chat Completions requests:

| Model | Use it for |
| - | - |
| `deepseek-v4-pro` | Everyday coding at a good price |
| `gpt-5.6-sol` | First choice when you want an OpenAI model |
| `qwen3-coder-flash` | Code generation and refactoring |
| `gemini-3.1-pro-preview` | Large codebases and long context |
| `deepseek-v4-flash`, `gpt-5.4-mini` | Small changes and low-cost tasks |

Prices are in the [model catalog](/en/models).

## FAQ

### With GPT-6 I get "Function tools with reasoning\_effort are not supported"

GPT-6 models can only call tools in Chat Completions with reasoning effort set to `none`; otherwise they need the Responses API. Cline's OpenAI Compatible provider uses Chat Completions, so switch to a model from the table above. To code with GPT-6, use [Codex CLI](/en/scenarios/programming/codex-cli) or [OpenCode](/en/scenarios/programming/opencode).

### I get "Unsupported parameter: 'max\_tokens'"

The model doesn't accept `max_tokens`, which is the case for GPT-6 models. Switch to a model from the table above.

### One task used a lot of tokens

Each step sends the conversation history, open files, and tool definitions. Give the token a quota and break large tasks into smaller ones. Models with a cache-read price, such as `deepseek-v4-pro`, bill repeated context at the cache rate; see [prompt cache billing](/en/faq/prompt-cache-billing).

## Related pages

* [Integrations overview](/en/scenarios)
* [OpenAI protocol: Chat Completions](/en/api-reference/chat-completions)
* [Cline docs: OpenAI Compatible](https://docs.cline.bot/provider-config/openai-compatible)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.