Authorizing with Cline

Connect Cline to an AI model. Sign in with a Cline account for instant access, or bring your own API key from Anthropic, OpenRouter, OpenAI, Google Gemini, or AWS Bedrock.

Cline needs a connection to an AI model before it can do any work. You have two paths:

Cline account

Sign in with Google, GitHub, or email. No API keys to manage. Access models from Anthropic, OpenAI, Google, and more through a single account with built-in billing.

Best for: Most users, fastest setup.

Bring your own key (BYOK)

Paste an API key from any supported provider: Anthropic, OpenRouter, OpenAI, Google Gemini, AWS Bedrock, and more. Run models locally with Ollama or LM Studio for full privacy.

Best for: Enterprise, custom billing, local models.

Sign in with a Cline account

The Cline account gives you access to multiple AI models through a single login — no separate API keys required.

1. Open the Cline panel

   Click the Cline icon in the Activity Bar (VS Code/Cursor) or go to View → Tool Windows → Cline (JetBrains).

2. Click Sign Up

   In the Cline panel, click Sign Up. Your browser opens to app.cline.bot.

3. Authenticate

   Choose Google, GitHub, or email to complete authentication.

4. Return to your editor

   After authentication, you are redirected back to your editor. Cline stores your session token in your IDE’s native secret storage (VS Code Secrets or JetBrains Credential Store). Your credentials are never sent to Cline’s servers.

5. Select a model

   Open Cline settings (gear icon in the Cline panel). Under API Provider, make sure Cline is selected. Then choose a model from the Model dropdown.

   To find no-cost options, search for free in the model selector — free models are labeled FREE.

6. Verify

   Send any message in the Cline chat. If Cline responds, you are ready to go.

Adding credits: Click Add Credits in Cline settings or visit your account dashboard. Credits work across all available models.

Bring your own API key (BYOK)

If you prefer to use your own API keys — for specific billing arrangements, higher rate limits, or access to beta models — select your provider in Cline settings and paste your key.

1. Open Cline settings

   Click the gear icon in the top-right of the Cline panel.

2. Choose a provider

   Select a provider from the API Provider dropdown.

3. Paste your API key

   Enter your API key in the field provided. Keys are stored in your system’s credential manager and sent only to your selected provider — never to Cline’s servers.

4. Select a model and verify

   Choose a model from the Model dropdown, then send a message to confirm the connection works.

Provider setup

Anthropic

Anthropic provides direct access to Claude models including Claude Sonnet and Claude Opus.

Get an API key:

1. Go to console.anthropic.com.
2. Sign in or create an account.
3. Navigate to API Keys and click Create Key.
4. Copy the key — it starts with sk-ant-.

Configure in Cline:

• API Provider: Anthropic
• API Key: Paste your key
• Model: Choose from available Claude models (Claude Sonnet 4 is recommended for reliability)

Claude Sonnet 4 offers the best balance of capability and speed for most coding tasks. Use Claude Opus for complex multi-step reasoning tasks.

OpenRouter

OpenRouter is a unified API gateway that provides access to dozens of models — Claude, GPT-4, Gemini, Llama, and more — through a single API key with competitive pricing.

Get an API key:

1. Go to openrouter.ai and sign in.
2. Navigate to Keys and click Create Key.
3. Copy the key — it starts with sk-or-.

Configure in Cline:

• API Provider: OpenRouter
• API Key: Paste your key
• Model: Cline fetches the latest model list from OpenRouter automatically, so new models appear as soon as they launch

OpenRouter is a good starting point if you want to experiment with different models without setting up multiple provider accounts.

OpenAI

OpenAI provides GPT-4o and other GPT models.

Get an API key:

1. Go to platform.openai.com and sign in.
2. Navigate to API Keys and click Create new secret key.
3. Copy the key — it starts with sk-.

Configure in Cline:

• API Provider: OpenAI
• API Key: Paste your key
• Model: Select from available GPT models

Cline also supports any OpenAI-compatible API endpoint. Select OpenAI Compatible as the provider, then enter a custom base URL to point to your endpoint.

Google Gemini

Google Gemini models offer very large context windows, useful for tasks involving large codebases or long documents.

Get an API key:

1. Go to aistudio.google.com and sign in with a Google account.
2. Click Get API key and create a key in a new or existing Google Cloud project.
3. Copy the key.

Configure in Cline:

• API Provider: Google Gemini
• API Key: Paste your key
• Model: Select from available Gemini models (Gemini 2.5 Pro has a very large context window)

Use Gemini models when you need to pass large amounts of code or documentation into the context window at once.

AWS Bedrock

AWS Bedrock provides managed access to Claude and other models within your AWS account. It is commonly used in enterprise environments with existing AWS infrastructure.

Prerequisites:

• An AWS account with Bedrock enabled in your region
• IAM credentials with bedrock:InvokeModel permissions
• Model access granted in the Bedrock console

Configure in Cline:

• API Provider: AWS Bedrock
• AWS Access Key ID: Your IAM access key
• AWS Secret Access Key: Your IAM secret key
• AWS Region: The region where Bedrock is enabled (e.g., us-east-1)
• Model: Select an available Bedrock model

You can also use IAM role-based authentication if you are running Cline in an environment that already has AWS credentials configured (for example, an EC2 instance with an attached role).

Local models

Run models on your own hardware for complete privacy and zero per-request cost.

Ollama

Easy setup with a wide model selection. Supports Llama, Qwen, Mistral, and many others. Run ollama serve and select Ollama in Cline settings.

LM Studio

GUI-based model management and download. Start the local server in LM Studio and select LM Studio as your provider in Cline settings.

Local models require sufficient hardware, particularly GPU memory. Smaller models (7B–14B parameters) work on consumer hardware; larger models need more RAM or a GPU. See the running models locally guide for requirements.

Choosing a model

Priority	Recommended model
Best overall reliability	Claude Sonnet 4
Best value	Qwen3 Coder
Fastest responses	Cerebras GLM 4.6
Full privacy	Any Ollama or LM Studio model
Existing Claude subscription	Claude Code (Max/Pro plan)

CLI authentication

If you use the Cline CLI, authenticate from your terminal:

cline auth

This opens a browser for OAuth authentication, the same flow as the IDE extension. Your session persists until you sign out.

Troubleshooting

Issue	Fix
”Unauthorized: Please sign in”	Session expired. Click Sign In in Cline settings to re-authenticate.
Browser doesn’t open for OAuth	Check your default browser settings. Copy the URL from the Cline output panel and open it manually.
Frequent re-authentication	Ensure you are not clearing IDE secrets. Try a full sign-out and sign-in.
API key not accepted	Verify the key is active and has not expired. Confirm you selected the matching provider in Cline settings.
Can’t access organization	Verify membership at app.cline.bot and ask your admin about permissions. Sign out and sign in again.

Next steps

Your first project

Build a working todo app with Cline to practice the core workflow.

Model selection guide

Compare models across speed, cost, and context window size to find the best fit for your tasks.