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

# LLM Providers

> Supported language model providers and configuration options for Tambo AI

Tambo supports multiple LLM providers, allowing you to choose the model that best fits your use case. You can configure providers through the Tambo Cloud dashboard or when self-hosting.

## Supported Providers

Tambo works with the following LLM providers:

### OpenAI

Supports GPT-4, GPT-4 Turbo, GPT-3.5 Turbo, and other OpenAI models.

```typescript theme={null}
// Configure via environment variable
OPENAI_API_KEY=sk-...
```

**Models:**

* `gpt-4o` - Latest GPT-4 Omni model
* `gpt-4-turbo` - Fast GPT-4 with 128k context
* `gpt-4` - Standard GPT-4
* `gpt-3.5-turbo` - Fast and cost-effective

### Anthropic

Supports Claude 3 family models including Opus, Sonnet, and Haiku.

```typescript theme={null}
// Configure via environment variable
ANTHROPIC_API_KEY=sk-ant-...
```

**Models:**

* `claude-3-opus` - Most capable Claude model
* `claude-3-sonnet` - Balanced performance and speed
* `claude-3-haiku` - Fast and cost-effective
* `claude-3-5-sonnet` - Latest Sonnet with enhanced capabilities

### Google Gemini

Supports Gemini Pro and other Google AI models.

```typescript theme={null}
// Configure via environment variable
GOOGLE_API_KEY=...
```

**Models:**

* `gemini-pro` - Google's most capable model
* `gemini-pro-vision` - Multimodal with image support
* `gemini-1.5-pro` - Extended context window

### Mistral AI

Supports Mistral's open and commercial models.

```typescript theme={null}
// Configure via environment variable
MISTRAL_API_KEY=...
```

**Models:**

* `mistral-large` - Most capable Mistral model
* `mistral-medium` - Balanced model
* `mistral-small` - Fast and efficient
* `mixtral-8x7b` - Open-source mixture of experts

### Cerebras

Ultra-fast inference with Cerebras hardware acceleration.

```typescript theme={null}
// Configure via environment variable
CEREBRAS_API_KEY=...
```

**Models:**

* `cerebras-gpt` - Fast inference on specialized hardware

### OpenAI-Compatible Providers

Tambo supports any provider with an OpenAI-compatible API, including:

* **Together AI** - Fast inference and fine-tuning
* **Anyscale** - Ray-based LLM serving
* **Replicate** - Cloud-based model hosting
* **Local models** via Ollama, LM Studio, etc.

```typescript theme={null}
// Configure custom endpoint
OPENAI_API_BASE=https://api.together.xyz/v1
OPENAI_API_KEY=...
```

## Configuration

### Tambo Cloud

When using Tambo Cloud, configure providers in your project settings:

1. Navigate to your project in the Tambo dashboard
2. Go to **Settings** → **LLM Providers**
3. Add your API keys for each provider
4. Select default model for your project

### Self-Hosted

When self-hosting, set environment variables in your `.env` file:

```bash theme={null}
# apps/api/.env
OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GOOGLE_API_KEY=...
MISTRAL_API_KEY=...
CEREBRAS_API_KEY=...

# Optional: Set fallback key
FALLBACK_OPENAI_API_KEY=sk-...
```

## Model Selection

### Per-Request Configuration

Specify model per request using the API:

```typescript theme={null}
const response = await tamboClient.messages.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "Hello!" }],
});
```

### Default Model

Set a default model in your project configuration or environment:

```bash theme={null}
DEFAULT_MODEL=gpt-4o
```

## Best Practices

### Model Selection

* **GPT-4o / Claude 3.5 Sonnet** - Best for complex reasoning and component generation
* **GPT-3.5 Turbo / Claude Haiku** - Fast and cost-effective for simple tasks
* **Gemini Pro** - Great for long context and multimodal tasks
* **Mistral** - Good balance of performance and cost
* **Cerebras** - Ultra-fast inference when speed is critical

### API Key Management

* Store keys securely in environment variables
* Never commit keys to version control
* Use different keys for development and production
* Rotate keys regularly
* Monitor usage and set spending limits

### Error Handling

```typescript theme={null}
try {
  const response = await tamboClient.messages.create({
    model: "gpt-4o",
    messages: [{ role: "user", content: "Hello!" }],
  });
} catch (error) {
  if (error.status === 401) {
    console.error("Invalid API key");
  } else if (error.status === 429) {
    console.error("Rate limit exceeded");
  } else {
    console.error("API error:", error.message);
  }
}
```

## Rate Limits

Each provider has different rate limits:

* **OpenAI**: Varies by tier (free, pay-as-you-go, enterprise)
* **Anthropic**: Based on API tier and usage
* **Google**: Requests per minute limit
* **Mistral**: Based on subscription plan
* **Cerebras**: Contact for limits

Implement retry logic with exponential backoff:

```typescript theme={null}
const maxRetries = 3;
let retries = 0;

while (retries < maxRetries) {
  try {
    return await tamboClient.messages.create(...);
  } catch (error) {
    if (error.status === 429 && retries < maxRetries - 1) {
      await new Promise(resolve => setTimeout(resolve, Math.pow(2, retries) * 1000));
      retries++;
    } else {
      throw error;
    }
  }
}
```

## Missing a Provider?

If you need support for a specific provider, please [open an issue on GitHub](https://github.com/tambo-ai/tambo/issues) or reach out on [Discord](https://discord.gg/dJNvPEHth6).

## Related Resources

* [Self-Hosting Guide](/deployment/self-hosting)
* [Environment Variables](/reference/environment-variables)
* [API Reference](/reference/api)