> ## Documentation Index > Fetch the complete documentation index at: https://fireworks.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Cursor IDE > Connect Cursor IDE to Fireworks models and Nexus Auto with the Nexus CLI or manual setup. Cursor CLI (`agent` or `cursor-agent`) and Cursor cloud agents aren't supported. Support summary: * API the harness speaks to Fireworks: OpenAI Chat Completions * Nexus Auto: Supported. Anthropic and OpenAI model turns need Provider Keys * Native web search: Not added by Nexus. See [Web search](https://fireworks.ai/docs/nexus/auto.md#web-search). Plan access: Pro and Team don't include general access to open models. They use Nexus Auto (`auto`) along with their own Claude or ChatGPT models, and Nexus Auto still sends turns to open models. Enterprise accounts also get general access to open models on Fireworks serverless. See [Plans and Accounts](https://fireworks.ai/docs/nexus/plans.md). ## Quick start **1. Install and sign in** ```bash wrap theme={null} curl -fsSL https://nexus.fireworks.ai/install.sh | bash ``` You need Node.js 22 or later, `git`, and `bash`. On Windows, run the installer from Git Bash. PowerShell breaks the script's line endings. ```bash wrap theme={null} nexus login ``` Sign in with your browser, or paste a Fireworks API key (`fw_...`). If the Nexus CLI is already installed and signed in, skip to step 2. **2. Quit Cursor IDE** > [!WARNING] > > **Fully quit Cursor IDE before you connect or run `off`.** A running Cursor IDE overwrites the change from memory. If Cursor IDE is open, Nexus CLI waits until you quit. `--force` writes anyway, which isn't recommended. **3. Connect Cursor IDE** ```bash wrap theme={null} nexus cursor ``` * Nexus CLI registers the Fireworks catalog in the picker and points every mode you already have in `modelConfig` at the default model. It never adds modes. * `nexus cursor status` shows the provider, where the key comes from, and the configured model. **4. Reopen and pick a model** Reopen Cursor IDE and pick **Auto** (`nexus/auto`), the default. To change the default for every mode, run `nexus cursor --model glm-latest`. ![Cursor IDE model picker showing Auto selected with Fireworks DeepSeek, GLM, Kimi, and MiniMax models](https://mintcdn.com/fireworksai-docs/VYpNGusqwFX7L4nT/images/nexus/cursor-ide-model-picker.png?fit=max\&auto=format\&n=VYpNGusqwFX7L4nT\&q=85\&s=62e6c6629cbb1328ca0d515053c595d2) Cursor IDE after Nexus CLI, with Auto selected (Enterprise account) While connected, only Fireworks models work. To restore your original settings and Cursor IDE's built-in models, run `nexus cursor off`. * Cursor IDE's built-in models are hidden while connected. * Cursor IDE also enforces a server-side allowlist, so not every Fireworks model is selectable. * Some features may still use Cursor's own backend, depending on plan and version. ## Use Nexus Auto * **Anthropic and OpenAI model turns** need a [Provider Key](https://fireworks.ai/docs/nexus/provider-keys.md) from your admin. Without one, Nexus Auto serves every turn with open models. - Nexus CLI registers Nexus Auto as `nexus/auto`, because Cursor reserves the bare `auto` name for its own router. - Cursor IDE can't send extra headers, so Nexus CLI doesn't send keys saved with `nexus auto`, and manual setup can't add them either. - `nexus auto --providers` has no effect in Cursor IDE. ## MCP and tools * Nexus CLI doesn't change your MCP servers. * Nexus CLI doesn't add [web search](https://fireworks.ai/docs/nexus/auto.md#web-search) to Cursor IDE. ## Troubleshooting * **`Model name is not valid: "auto"`, or the same error for another Fireworks ID:** Cursor IDE stopped sending your Fireworks key. Open **Settings → Models**, turn **OpenAI API Key** back on, and retry. * **The key was revoked:** run `nexus login --force` for a new key, fully quit Cursor IDE, and run `nexus cursor`. - Cursor can turn the key off after Fireworks rejects a request, such as when the key is revoked or a model ID is not found. Without a key, Cursor treats Fireworks IDs as unknown models. - For a revoked key, turning it back on is not enough. With manual setup, paste a new key from [Nexus settings](https://nexus.fireworks.ai/settings/api-keys). - See [Nexus Auto](https://fireworks.ai/docs/nexus/auto.md#troubleshooting) for issues that affect every harness. ## Manual setup (advanced) In Cursor IDE, open **Settings → Models**: 1. Under **API Keys**, paste your Fireworks key into **OpenAI API Key**. 2. Turn on **Override OpenAI Base URL** and enter `https://api.fireworks.ai/inference/v1`. 3. Add a custom model with a Fireworks ID, such as `glm-latest`, then select it in chat. Without the Nexus CLI, you take over [the tasks it otherwise performs](https://fireworks.ai/docs/nexus/manual-setup.md#what-changes-without-the-nexus-cli), such as undo and updates. ## What the Nexus CLI writes Cursor IDE stores AI settings in SQLite, in `User/globalStorage/state.vscdb` under `~/.config/Cursor` (Linux), `~/Library/Application Support/Cursor` (macOS), or `%APPDATA%\Cursor` (Windows): | Setting | Value | | - | - | | `secret://cursorAuth/openAIKey` | Your Fireworks key, encrypted with Electron `safeStorage` | | `cursorAuth/openAIKey` | The same key in plain text, for older Cursor builds | | `openAIBaseUrl` | `https://api.fireworks.ai/inference/v1` | | `useOpenAIKey` | `true` | | `aiSettings.userAddedModels` | The Fireworks catalog, tracked for a clean `off` | | `aiSettings.modelOverrideDisabled` | Cursor's built-in models, hidden while connected | | `aiSettings.modelConfig[]` | The Fireworks model for each mode | * On Linux, encryption needs `libsecret`. Without it, Cursor IDE only obfuscates the key and Nexus CLI warns you. * `off` restores your previous AI settings and key from the snapshot under `~/.nexus/cursor/`. Without a snapshot, it removes only what Nexus CLI added. * Use `--db-path` for a non-default `state.vscdb`. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.