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

# Cloud Providers

> Configure Cline to use Anthropic, OpenRouter, OpenAI, Google Gemini, AWS Bedrock, Azure OpenAI, and GCP Vertex AI.

Cline connects to over 30 cloud providers. This page covers the major ones: what credentials you need, where to get them, and how to configure each in Cline.

<Tip>
  If you're unsure which provider to start with, **OpenRouter** gives you access to models from all major providers through a single API key — useful for experimenting before committing to a single provider.
</Tip>

## Configuring a provider

All providers are configured in the same place:

1. Click the settings gear icon (⚙️) in the Cline panel.
2. Select your provider from the **API Provider** dropdown.
3. Enter your API key or credentials.
4. Choose a model from the **Model** dropdown.

***

## Provider setup

<Tabs>
  <Tab title="Anthropic">
    Anthropic makes the Claude family of models. Claude Sonnet and Opus have the strongest tool-use reliability of any model family, making Anthropic a popular first choice for Cline users.

    **Website:** [anthropic.com](https://www.anthropic.com)

    ### Getting an API key

    <Steps>
      <Step title="Sign in to Anthropic Console">
        Go to [console.anthropic.com](https://console.anthropic.com) and create an account or sign in.
      </Step>

      <Step title="Navigate to API keys">
        Go to [console.anthropic.com/settings/keys](https://console.anthropic.com/settings/keys).
      </Step>

      <Step title="Create a key">
        Click **Create Key**, give it a descriptive name (e.g., `cline`), and click **Create Key**.
      </Step>

      <Step title="Copy the key immediately">
        Copy your key now — Anthropic will not show it again. Store it in a password manager or secrets vault.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `Anthropic`.
    2. Paste your key into the **Anthropic API Key** field.
    3. Select a model — `claude-sonnet-4-5-20250929` is the recommended default.

    ### Available models

    | Model                           | Context | Notes                  |
    | ------------------------------- | ------- | ---------------------- |
    | `claude-sonnet-4-5-20250929`    | 200K    | Recommended default    |
    | `claude-sonnet-4-5-20250929:1m` | 1M      | Large codebase variant |
    | `claude-opus-4-20250514`        | 200K    | Most capable           |
    | `claude-opus-4-20250514:1m`     | 1M      | Large codebase variant |
    | `claude-haiku-4-5-20251001`     | 200K    | Fast, affordable       |
    | `claude-sonnet-4-6`             | 200K    | Latest Sonnet          |

    ### Notes

    * **Prompt caching** is enabled automatically and reduces costs significantly for repeated context.
    * **Extended thinking** is available on Sonnet 3.7+ and Opus 4 — enable it in Cline settings for complex reasoning tasks.
    * **Rate limits** are tier-based. If you hit limits frequently, consider accessing Claude through [OpenRouter](#openrouter) instead.
    * **Pricing:** [anthropic.com/pricing](https://www.anthropic.com/pricing)
  </Tab>

  <Tab title="OpenRouter">
    OpenRouter is a unified API gateway that gives you access to models from Anthropic, OpenAI, Google, Mistral, Meta, DeepSeek, and many others — all through a single API key. Cline automatically fetches the full model list from OpenRouter.

    **Website:** [openrouter.ai](https://openrouter.ai)

    ### Getting an API key

    <Steps>
      <Step title="Sign in to OpenRouter">
        Go to [openrouter.ai](https://openrouter.ai) and sign in with your Google or GitHub account.
      </Step>

      <Step title="Get your API key">
        Navigate to [openrouter.ai/keys](https://openrouter.ai/keys). Your default key is shown there. If none exists, click **Create Key**.
      </Step>

      <Step title="Add credits">
        OpenRouter requires prepaid credits. Add credits at [openrouter.ai/credits](https://openrouter.ai/credits). Usage is billed against the underlying model's pricing.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `OpenRouter`.
    2. Paste your key into the **OpenRouter API Key** field.
    3. Select any model from the dropdown — the list is fetched automatically.

    ### Notes

    * **Model selection:** OpenRouter hosts 100+ models. See [openrouter.ai/models](https://openrouter.ai/models) for the full list with pricing.
    * **Prompt compression:** Enable "Compress prompts and message chains to the context size" in Cline to activate OpenRouter's middle-out transform, which helps when prompts approach the model's context limit.
    * **Prompt caching:** Automatically passed through to models that support it. For Gemini models specifically, you must manually enable **Enable Prompt Caching** in Cline's provider settings.
    * **Rate limits:** OpenRouter distributes requests across multiple providers, so you're less likely to hit per-provider rate limits.
    * **Pricing:** [openrouter.ai/models](https://openrouter.ai/models)
  </Tab>

  <Tab title="OpenAI">
    OpenAI provides the GPT-4 and GPT-5 model families, plus reasoning-focused o-series models. If you already have an OpenAI account, this is the most direct way to access GPT models.

    **Website:** [openai.com](https://openai.com)

    ### Getting an API key

    <Steps>
      <Step title="Sign in to OpenAI Platform">
        Go to [platform.openai.com](https://platform.openai.com) and sign in.
      </Step>

      <Step title="Navigate to API keys">
        Go to [platform.openai.com/api-keys](https://platform.openai.com/api-keys).
      </Step>

      <Step title="Create a key">
        Click **Create new secret key**, give it a name like `cline`, and click **Create secret key**.
      </Step>

      <Step title="Copy the key immediately">
        Copy your key now — OpenAI will not show it again.
      </Step>

      <Step title="Add billing">
        Go to [platform.openai.com/account/billing](https://platform.openai.com/account/billing) and add a payment method. Usage is billed monthly.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `OpenAI`.
    2. Paste your key into the **OpenAI API Key** field.
    3. Select a model from the dropdown.

    ### Available models (selected)

    | Model         | Context | Notes              |
    | ------------- | ------- | ------------------ |
    | `gpt-5.2`     | 400K    | Current flagship   |
    | `gpt-4o`      | 128K    | Fast, multimodal   |
    | `gpt-4o-mini` | 128K    | Affordable, fast   |
    | `gpt-4.1`     | 1M+     | Large context      |
    | `o3`          | 200K    | Advanced reasoning |
    | `o4-mini`     | 200K    | Fast reasoning     |

    ### Notes

    * **ChatGPT subscription:** If you have a ChatGPT Plus, Pro, or Team plan, use the **OpenAI Codex** provider instead to use your subscription directly — no separate API billing needed.
    * **Azure OpenAI:** For enterprise Azure deployments, see the Azure tab.
    * **Pricing:** [openai.com/pricing](https://openai.com/pricing)
  </Tab>

  <Tab title="Google Gemini">
    Google's Gemini models offer some of the largest context windows available — up to 2M tokens — making them well-suited for large codebases and long documents.

    **Website:** [ai.google.dev](https://ai.google.dev)

    ### Getting an API key

    <Steps>
      <Step title="Sign in to Google AI Studio">
        Go to [aistudio.google.com](https://aistudio.google.com) and sign in with your Google account.
      </Step>

      <Step title="Create an API key">
        Navigate to [aistudio.google.com/apikey](https://aistudio.google.com/apikey) and click **Create API Key**. Select an existing Google Cloud project or create a new one.
      </Step>

      <Step title="Copy the key">
        Copy the API key and store it securely.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `Google Gemini`.
    2. Paste your key into the **Gemini API Key** field.
    3. Select a model from the dropdown.

    ### Available models (selected)

    | Model                    | Context | Price (input/M tokens) | Notes             |
    | ------------------------ | ------- | ---------------------- | ----------------- |
    | `gemini-2.5-pro`         | 1M      | $1.25–$2.50            | Best performance  |
    | `gemini-2.5-flash`       | 1M      | \$0.30                 | Fast, affordable  |
    | `gemini-3.1-pro-preview` | 1M      | $2.00–$4.00            | Latest generation |
    | `gemini-2.0-flash-001`   | 1M      | \$0.10                 | Budget option     |
    | `gemini-1.5-pro-002`     | 2M      | Tiered                 | Largest context   |

    ### Notes

    * **Thinking/reasoning:** Gemini 2.5 and 3 series support thinking budgets and levels. Enable extended thinking in Cline for complex tasks.
    * **Tiered pricing:** Some models charge less per token under 200K tokens. For most coding tasks, you'll benefit from the lower tier.
    * **Free experimental models:** Several `exp` variants are available at no cost during their preview period.
    * **Prompt caching:** Available on Gemini 2.5+ and select 2.0 models. When using Gemini via OpenRouter, you must manually enable prompt caching in Cline.
    * **Pricing:** [ai.google.dev/gemini-api/docs/models/gemini](https://ai.google.dev/gemini-api/docs/models/gemini)
  </Tab>

  <Tab title="AWS Bedrock">
    AWS Bedrock gives you access to Claude, Llama, Mistral, and other models through your AWS account. Useful for teams that need to keep API traffic within AWS infrastructure or want consolidated AWS billing.

    **Website:** [aws.amazon.com/bedrock](https://aws.amazon.com/bedrock)

    ### Getting credentials

    Bedrock uses AWS IAM credentials rather than a simple API key.

    <Steps>
      <Step title="Enable Bedrock model access">
        In the AWS Console, go to **Amazon Bedrock → Model access** and request access to the models you want to use (e.g., Anthropic Claude). Approval is usually instant.
      </Step>

      <Step title="Create an IAM user or role">
        Go to **IAM → Users** and create a user (or use an existing role). Attach the `AmazonBedrockFullAccess` policy or a scoped policy that includes `bedrock:InvokeModel`.
      </Step>

      <Step title="Generate access keys">
        For IAM users: go to the user's **Security credentials** tab and click **Create access key**. Download the access key ID and secret access key.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `AWS Bedrock`.
    2. Enter your **AWS Access Key ID** and **AWS Secret Access Key**.
    3. Set your **AWS Region** (e.g., `us-east-1`). Bedrock model availability varies by region.
    4. Select your model.

    <Note>
      You can also use environment variables (`AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, `AWS_REGION`) or an AWS profile if you prefer not to store credentials in Cline settings.
    </Note>

    ### Notes

    * **Region availability:** Not all models are available in all regions. Check [AWS Bedrock model availability](https://docs.aws.amazon.com/bedrock/latest/userguide/models-regions.html) before selecting a region.
    * **Pricing:** You pay AWS directly based on token usage. Prices are roughly equivalent to going direct, sometimes with a small markup. See [aws.amazon.com/bedrock/pricing](https://aws.amazon.com/bedrock/pricing).
  </Tab>

  <Tab title="Azure OpenAI">
    Azure OpenAI gives enterprise access to OpenAI models through Microsoft Azure infrastructure, with data residency options and Azure AD authentication.

    **Website:** [azure.microsoft.com/products/ai-services/openai-service](https://azure.microsoft.com/products/ai-services/openai-service)

    ### Getting credentials

    Azure OpenAI requires a deployment rather than selecting a model directly.

    <Steps>
      <Step title="Create an Azure OpenAI resource">
        In the [Azure Portal](https://portal.azure.com), create an **Azure OpenAI** resource in a supported region.
      </Step>

      <Step title="Deploy a model">
        In **Azure AI Foundry** (formerly Azure OpenAI Studio), create a deployment for a model (e.g., `gpt-4o`). Note the **deployment name** you choose — this is what you'll enter in Cline.
      </Step>

      <Step title="Get your endpoint and API key">
        In the Azure Portal, go to your Azure OpenAI resource → **Keys and Endpoint**. Copy your **Endpoint URL** and **Key 1**.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `Azure OpenAI`.
    2. Enter your **Azure OpenAI API Key**.
    3. Enter your **Endpoint URL** (e.g., `https://your-resource.openai.azure.com`).
    4. Enter your **Deployment Name** (the name you gave your model deployment).
    5. Set the **API Version** (e.g., `2024-12-01-preview`).

    ### Notes

    * **Model versions:** Azure OpenAI deploys specific model versions, which may lag behind the versions available on OpenAI's direct API.
    * **Pricing:** [azure.microsoft.com/pricing/details/cognitive-services/openai-service](https://azure.microsoft.com/pricing/details/cognitive-services/openai-service)
  </Tab>

  <Tab title="GCP Vertex AI">
    GCP Vertex AI gives you access to Claude (via Anthropic's Vertex integration) and Google's Gemini models within your Google Cloud infrastructure.

    **Website:** [cloud.google.com/vertex-ai](https://cloud.google.com/vertex-ai)

    ### Getting credentials

    Vertex AI uses Google Cloud service account credentials.

    <Steps>
      <Step title="Enable the Vertex AI API">
        In the [Google Cloud Console](https://console.cloud.google.com), navigate to **APIs & Services → Enable APIs** and enable the **Vertex AI API**.
      </Step>

      <Step title="Enable model access">
        For Claude on Vertex: go to **Vertex AI → Model Garden** and click **Enable** for the Claude model(s) you want.
      </Step>

      <Step title="Create a service account">
        Go to **IAM & Admin → Service Accounts** and create a service account. Grant it the **Vertex AI User** role (`roles/aiplatform.user`).
      </Step>

      <Step title="Download a key">
        On the service account, go to **Keys → Add Key → Create new key** and download a JSON key file.
      </Step>
    </Steps>

    ### Configuration in Cline

    1. Set **API Provider** to `GCP Vertex AI`.
    2. Enter your **Google Cloud Project ID**.
    3. Enter your **Region** (e.g., `us-east5`).
    4. Provide your service account credentials — either paste the JSON key contents or set the `GOOGLE_APPLICATION_CREDENTIALS` environment variable to the path of your key file.
    5. Select your model.

    ### Notes

    * **Supported models:** Vertex AI supports Gemini models natively and Claude models through Anthropic's Vertex partnership.
    * **Region:** Claude on Vertex is available in limited regions (e.g., `us-east5`). Check [Anthropic's Vertex documentation](https://docs.anthropic.com/en/api/claude-on-vertex-ai) for current availability.
    * **Pricing:** [cloud.google.com/vertex-ai/pricing](https://cloud.google.com/vertex-ai/pricing)
  </Tab>
</Tabs>

***

## Which provider should I use?

| Situation                        | Recommended provider              |
| -------------------------------- | --------------------------------- |
| Getting started, want simplicity | OpenRouter (one key, many models) |
| Best Claude reliability          | Anthropic direct                  |
| Best GPT models                  | OpenAI direct                     |
| Largest context window           | Google Gemini                     |
| Enterprise AWS infrastructure    | AWS Bedrock                       |
| Enterprise Azure infrastructure  | Azure OpenAI                      |
| Enterprise GCP infrastructure    | GCP Vertex AI                     |
| Access to many models at once    | OpenRouter                        |

## Next steps

<CardGroup cols={2}>
  <Card title="Model selection guide" icon="list-check" href="/models/model-selection-guide">
    Compare models across providers and find the right one for your task.
  </Card>

  <Card title="Context windows" icon="layers" href="/models/context-windows">
    Understand how context limits affect large codebases and how Cline manages them.
  </Card>
</CardGroup>