> ## Documentation Index
> Fetch the complete documentation index at: https://thethirdpenco-feat-tool-lifecycle-events.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# DeepSeek

> Use DeepSeek's advanced AI models

DeepSeek provides powerful AI models including their flagship chat model and the DeepSeek Reasoner for complex reasoning tasks. Their API is OpenAI-compatible, making integration seamless.

## Setup

Set your API key as an environment variable:

```bash theme={null}
export DEEPSEEK_API_KEY="sk-..."
```

Get your API key from [DeepSeek Platform](https://platform.deepseek.com/).

## Usage

```python theme={null}
from ai_query import generate_text
from ai_query.providers import deepseek

result = await generate_text(
    model=deepseek("deepseek-chat"),
    prompt="Explain machine learning in simple terms."
)
```

## Available Models

| Model ID            | Description                              |
| ------------------- | ---------------------------------------- |
| `deepseek-chat`     | DeepSeek-V3 - general purpose chat model |
| `deepseek-reasoner` | DeepSeek-R1 - advanced reasoning model   |

## Provider Options

Customize parameters:

````python theme={null}
result = await generate_text(
    model=deepseek("deepseek-chat"),
    prompt="Write a creative story.",
    provider_options={
        "deepseek": {
            "temperature": 0.9,
            "max_tokens": 1000
        }
    }
)

## Using DeepSeek Reasoner

DeepSeek Reasoner (R1) excels at complex reasoning tasks:

```python
from ai_query import generate_text
from ai_query.providers import deepseek

result = await generate_text(
    model=deepseek("deepseek-reasoner"),
    prompt="Solve this step by step: If a train travels 120 km in 2 hours, then stops for 30 minutes, then travels another 90 km in 1.5 hours, what is its average speed for the entire journey?"
)
````

## Tool Calling

Tool calling works the same as with other providers:

```python theme={null}
from ai_query import generate_text
from ai_query.providers import deepseek, tool, Field

@tool(description="Calculate a math expression")
async def calculate(expression: str = Field(description="Math expression")) -> str:
    return str(eval(expression))

result = await generate_text(
    model=deepseek("deepseek-chat"),
    prompt="What is 25 * 17 + 89?",
    tools={"calculate": calculate}
)
```

## Streaming

```python theme={null}
from ai_query import stream_text
from ai_query.providers import deepseek

result = stream_text(
    model=deepseek("deepseek-chat"),
    prompt="Write a poem about coding."
)

async for chunk in result.text_stream:
    print(chunk, end="", flush=True)
```

## Why DeepSeek?

* **Cost effective**: Highly competitive pricing
* **Strong performance**: Excellent benchmark results
* **Reasoning capabilities**: DeepSeek-R1 excels at complex reasoning
* **OpenAI compatible**: Easy migration from OpenAI