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

# Providers

> OpenAI, Anthropic, Google, and Amazon Bedrock provider setup and configuration

Tracia uses the [Vercel AI SDK](https://sdk.vercel.ai/docs) under the hood. Install only the providers you need.

## Installation

Install the core AI SDK package plus your provider(s):

```bash theme={null}
# Core package (required)
npm install ai

# Provider packages (install the ones you need)
npm install @ai-sdk/openai          # For OpenAI models
npm install @ai-sdk/anthropic       # For Anthropic models
npm install @ai-sdk/google          # For Google models
npm install @ai-sdk/amazon-bedrock  # For Amazon Bedrock models
```

## OpenAI

### Environment Variable

```bash theme={null}
OPENAI_API_KEY=sk-your-openai-key
```

### Usage

```typescript theme={null}
const result = await tracia.runLocal({
  model: 'gpt-4o',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello!' }
  ],
  temperature: 0.7,
  maxOutputTokens: 500
});
```

### Streaming

```typescript theme={null}
const stream = tracia.runLocal({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Write a poem.' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk);
}
```

## Anthropic

### Environment Variable

```bash theme={null}
ANTHROPIC_API_KEY=sk-ant-your-anthropic-key
```

### Usage

```typescript theme={null}
const result = await tracia.runLocal({
  model: 'claude-sonnet-4-20250514',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Explain quantum computing.' }
  ],
  temperature: 0.7,
  maxOutputTokens: 1000
});
```

### Streaming

```typescript theme={null}
const stream = tracia.runLocal({
  model: 'claude-sonnet-4-20250514',
  messages: [{ role: 'user', content: 'Write a story.' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk);
}
```

## Google

### Environment Variable

```bash theme={null}
GOOGLE_API_KEY=your-google-api-key
```

### Usage

```typescript theme={null}
const result = await tracia.runLocal({
  model: 'gemini-2.0-flash',
  messages: [
    { role: 'user', content: 'What are the benefits of TypeScript?' }
  ],
  temperature: 0.5,
  maxOutputTokens: 800
});
```

### Streaming

```typescript theme={null}
const stream = tracia.runLocal({
  model: 'gemini-2.0-flash',
  messages: [{ role: 'user', content: 'Explain machine learning.' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk);
}
```

## Provider Override

The SDK automatically detects the provider based on the model name. For custom or fine-tuned models not in the built-in list, specify the provider explicitly:

```typescript theme={null}
// Fine-tuned OpenAI model
const result = await tracia.runLocal({
  model: 'ft:gpt-4o:my-org:custom-model:abc123',
  provider: 'openai',
  messages: [{ role: 'user', content: 'Hello!' }]
});
```

## API Key Override

Override the environment variable for a specific request:

```typescript theme={null}
// Use a different API key
const result = await tracia.runLocal({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Hello!' }],
  providerApiKey: process.env.OPENAI_API_KEY_SECONDARY
});
```

This is useful for:

* Multi-tenant applications with per-customer API keys
* Testing with different API keys
* Using organization-specific keys

## Error Handling by Provider

```typescript theme={null}
import { Tracia, TraciaError, TraciaErrorCode } from 'tracia';

try {
  const result = await tracia.runLocal({
    model: 'gpt-4o',
    messages: [{ role: 'user', content: 'Hello!' }]
  });
} catch (error) {
  if (error instanceof TraciaError) {
    switch (error.code) {
      case TraciaErrorCode.MISSING_PROVIDER_SDK:
        // Missing ai package or provider package
        console.error('Install required packages: npm install ai @ai-sdk/openai');
        break;
      case TraciaErrorCode.MISSING_PROVIDER_API_KEY:
        console.error('Set OPENAI_API_KEY environment variable');
        break;
      case TraciaErrorCode.PROVIDER_ERROR:
        console.error('Provider API error:', error.message);
        break;
    }
  }
}
```

## Provider Comparison

| Feature           | OpenAI          | Anthropic       | Google               | Amazon Bedrock   |
| ----------------- | --------------- | --------------- | -------------------- | ---------------- |
| System messages   | Native support  | Native support  | Converted internally | Vendor-dependent |
| Temperature range | 0-2             | 0-1             | 0-2                  | Vendor-dependent |
| Max output tokens | Model-dependent | Model-dependent | Model-dependent      | Model-dependent  |
| Stop sequences    | Supported       | Supported       | Supported            | Supported        |
| Streaming         | Supported       | Supported       | Supported            | Supported        |
| Tool calling      | Supported       | Supported       | Supported            | Model-dependent  |

<Note>
  Amazon Bedrock hosts models from multiple vendors (Amazon, Anthropic, Meta, Mistral, etc.). Feature support depends on the specific model. See the [Bedrock page](/sdk-node/run-local/bedrock) for details.
</Note>