> ## Documentation Index
> Fetch the complete documentation index at: https://0g.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Create

# Create Chat

Create a new chat instance to interact with AI models.

## Constructor

```typescript theme={null}
new Chat(config: ChatConfig)
```

### Parameters

<ParamField path="config" type="ChatConfig" required>
  Configuration object for the chat instance
</ParamField>

<ParamField path="config.apiKey" type="string" required>
  Your API key for authentication
</ParamField>

<ParamField path="config.model" type="string" default="gpt-3.5-turbo">
  The AI model to use for chat completions
</ParamField>

<ParamField path="config.maxTokens" type="number" default="1000">
  Maximum number of tokens in the response
</ParamField>

<ParamField path="config.temperature" type="number" default="0.7">
  Controls randomness in responses (0-1)
</ParamField>

<ParamField path="config.timeout" type="number" default="30000">
  Request timeout in milliseconds
</ParamField>

<ParamField path="config.baseURL" type="string">
  Custom base URL for the API endpoint
</ParamField>

## Response

Returns a `Chat` instance that can be used to send messages and stream responses.

## Example

<CodeGroup>
  ```typescript Basic Usage theme={null}
  import { createAgent } from '@src/index';

  const agent = await createAgent({
    name: 'My Assistant',
    providerAddress: '0xf07240Efa67755B5311bc75784a061eDB47165Dd', // llama-3.3-70b-instruct
    memoryBucket: 'my-agent-memory',
    privateKey: 'your-private-key',
    maxTokens: 2000,
    temperature: 0.8
  });
  ```

  ```typescript Advanced Reasoning Model theme={null}
  import { createAgent } from '@src/index';

  const agent = await createAgent({
    name: 'Reasoning Assistant',
    providerAddress: '0x3feE5a4dd5FDb8a32dDA97Bed899830605dBD9D3', // deepseek-r1-70b
    memoryBucket: 'reasoning-memory',
    privateKey: 'your-private-key',
    maxTokens: 1500,
    temperature: 0.3
  });
  ```

  ```typescript Custom Configuration theme={null}
  import { createAgent } from '@src/index';

  const agent = await createAgent({
    name: 'Custom Assistant',
    providerAddress: '0xf07240Efa67755B5311bc75784a061eDB47165Dd',
    memoryBucket: 'custom-memory',
    privateKey: 'your-private-key',
    rpcUrl: 'https://custom-rpc.0g.ai',
    indexerRpc: 'https://custom-indexer.0g.ai',
    kvRpc: 'https://custom-kv.0g.ai',
    maxEphemeralMessages: 100,
    temperature: 0.5,
    maxTokens: 500
  });
  ```
</CodeGroup>

## Available Models

The 0G AI SDK connects to models running on the 0G decentralized compute network:

| Model                    | Provider Address                             | Description                                                    | Verification |
| ------------------------ | -------------------------------------------- | -------------------------------------------------------------- | ------------ |
| `llama-3.3-70b-instruct` | `0xf07240Efa67755B5311bc75784a061eDB47165Dd` | State-of-the-art 70B parameter model for general AI tasks      | TEE (TeeML)  |
| `deepseek-r1-70b`        | `0x3feE5a4dd5FDb8a32dDA97Bed899830605dBD9D3` | Advanced reasoning model optimized for complex problem solving | TEE (TeeML)  |

## Error Handling

The constructor will throw an error if:

* No API key is provided and `ZG_API_KEY` environment variable is not set
* Invalid configuration parameters are passed

```typescript theme={null}
try {
  const chat = new Chat({
    apiKey: 'invalid-key',
    model: 'llama-3.3-70b-instruct'
  });
} catch (error) {
  console.error('Failed to create chat:', error.message);
}
```

## Next Steps

* [Send Messages](/api-reference/chat/stream)
* [Stream Responses](/api-reference/chat/stream)
* [View Chat History](/api-reference/chat/history)