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

# Claude Code

[Claude Code](https://claude.com/claude-code) is Anthropic's developer toolkit for working on a local codebase from the terminal or inside VS Code. It can chat, edit files, and run commands with your approval. Because Claude Code uses the Anthropic API, you can point it to the ScitiX Model Inference API and use supported ScitiX models in the same workflow.

Currently supported models are `Qwen/Qwen3.5-397B-A17B`, `MiniMaxAI/MiniMax-M2.5`, `MiniMaxAI/MiniMax-M2.7`, and `kimi-k2.6`. For model pricing, see the [model catalog](https://console.scitix.ai/model-inference/models).

This guide covers both the command-line tool and the VS Code extension.

## Configure the CLI

### 1. Install Claude Code

Install the CLI by following [Anthropic's official instructions](https://claude.com/claude-code).

### 2. Configure environment variables

Create an API Key on the [API Keys](https://console.scitix.ai/model-inference/api_keys) page, then set the following environment variables:

```bash theme={null}
export ANTHROPIC_BASE_URL="https://api.scitix.ai/model-api"
export ANTHROPIC_AUTH_TOKEN="<Your API Key>"
```

To keep them across sessions on macOS or Linux, add these lines to `~/.zshrc` or `~/.bashrc` and restart the terminal.

Keep your API Key secure, and never commit it to Git. Environment variables are recommended.

Optionally, set a fast model for lightweight background tasks:

```bash theme={null}
export ANTHROPIC_DEFAULT_HAIKU_MODEL="claude-haiku-4-5-20251001"
```

The older `ANTHROPIC_SMALL_FAST_MODEL` variable is deprecated. Use `ANTHROPIC_DEFAULT_HAIKU_MODEL` instead.

### 3. Run Claude Code

Run Claude Code from your project directory and pass the model name exactly as it appears on the platform:

```bash theme={null}
claude --model Qwen/Qwen3.5-397B-A17B
```

## Configure the VS Code Extension

### 1. Install the extension

Open **Extensions** in VS Code (`Ctrl+Shift+X` / `Cmd+Shift+X`), search for **claude**, and install **Claude Code for VS Code** by Anthropic.

<img src="https://mintcdn.com/scitix-adfc65c2/zz3UuyM3SCvVbeP9/model-inference/media/claude-install.png?fit=max&auto=format&n=zz3UuyM3SCvVbeP9&q=85&s=12b7737e174426f74bfb256df554a3d5" alt="Claude Code — install from the Extensions marketplace" width="2904" height="1468" data-path="model-inference/media/claude-install.png" />

### 2. Configure the connection

Create or edit `~/.claude/settings.json`, then add the configuration below. Replace `<API_KEY>` with your ScitiX API Key.

```json theme={null}
{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.scitix.ai/model-api",
    "ANTHROPIC_AUTH_TOKEN": "<API_KEY>",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001",
    "CLAUDE_CODE_ENABLE_TELEMETRY": "0",
    "CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC": "1",
    "CLAUDE_CODE_ATTRIBUTION_HEADER": "0"
  },
  "prefersReducedMotion": true,
  "model": "Qwen/Qwen3.5-397B-A17B"
}
```

The fields that connect Claude Code to ScitiX are:

| Field                               | Description                                                                                                                                      |
| ----------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ |
| `env.ANTHROPIC_BASE_URL`            | ScitiX Model Inference API base URL: `https://api.scitix.ai/model-api`                                                                           |
| `env.ANTHROPIC_AUTH_TOKEN`          | Your API Key from the [API Keys](https://console.scitix.ai/model-inference/api_keys) page.                                                       |
| `env.ANTHROPIC_DEFAULT_HAIKU_MODEL` | Optional fast model for lightweight background tasks, such as `claude-haiku-4-5-20251001`. Replaces the deprecated `ANTHROPIC_SMALL_FAST_MODEL`. |
| `model`                             | The model to use, such as `Qwen/Qwen3.5-397B-A17B`.                                                                                              |

After saving, reload the VS Code window so the settings take effect.

<img src="https://mintcdn.com/scitix-adfc65c2/zz3UuyM3SCvVbeP9/model-inference/media/claude-settings.png?fit=max&auto=format&n=zz3UuyM3SCvVbeP9&q=85&s=00cbba7ad9a195f38fb4230ac77ea69a" alt="Claude Code — configure the connection in settings.json" width="3002" height="1776" data-path="model-inference/media/claude-settings.png" />

### 3. Choose an edit mode

Click the Claude Code icon in the top-right of VS Code to open the panel. From the bottom-left of the panel, choose how edits are applied.

| Mode                   | When to use                                   |
| ---------------------- | --------------------------------------------- |
| **Ask before edits**   | Review each change before it is applied.      |
| **Edit automatically** | Apply edits directly for quick iterations.    |
| **Plan mode**          | Claude outlines a plan before making changes. |