> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.reka.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.reka.ai/_mcp/server.

# Quickstart

> Start using Reka Research in minutes — with OpenAI-compatible calls, real-time web search, and grounded answers.

This page shows you the steps needed to get started with the Reka API and running the Reka Edge model locally.

* Run models on the Reka Platform
* Run Reka Edge locally (macOS)
* Run Reka Edge locally using OpenAI-compatible server (Linux)

## Run models on the Reka Platform

Create a free account on the [Reka Platform](https://platform.reka.ai/) to access your API key.

Keep your API key secure. Never expose it in client-side code or share it publicly.

### Install the SDK

Install the [OpenAI Python SDK](https://pypi.org/project/openai/) — our API is fully OpenAI-compatible.

```bash
pip install openai
```

### Available models

Our baseline models always available for public access are:

| model name                    |
| ----------------------------- |
| reka-flash                    |
| reka-edge (or reka-edge-2603) |

### Make your first request

```python "First request"
from openai import OpenAI

client = OpenAI(
    base_url="https://api.reka.ai/v1",
    api_key="YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="reka-edge",  # or "reka-flash"
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "image_url", "image_url": {"url": "https://v0.docs.reka.ai/_images/000000245576.jpg"}},
                {"type": "text", "text": "What do you like about this image?"}
            ],
        }
    ],
)
print(response.choices[0].message.content)
```

## Run Reka Edge locally (macOS)

See our [HuggingFace repository](https://huggingface.co/RekaAI/reka-edge-2603) for instructions on running Reka Edge locally.

### Requirements

* **OS:** macOS 13+
* **Hardware:** Apple Silicon Mac with 32 GB+ unified memory (M1 Pro/Max or later recommended)
* **Python:** 3.12+
* [**uv**](https://docs.astral.sh/uv/) (recommended) — handles dependencies automatically

## Run Reka Edge locally using OpenAI-compatible server (Linux)

For high-throughput serving, you can use the [vllm-reka](https://github.com/reka-ai/vllm-reka) plugin that extends standard [vLLM](https://github.com/vllm-project/vllm) to support Reka's custom architectures and optimized tokenizer. Please follow our [vllm-reka installation instructions](https://github.com/reka-ai/vllm-reka/blob/main/README.md) to install the plugin along with `vLLM`.

### Requirements

* **OS**: Linux with CUDA. macOS is not supported for serving.
* **Hardware**: NVIDIA GPU, ideally with ≥24 GB VRAM. This has been tested to work on GTX 3090 GPUs with 40-50 tokens/s.
* **Python**: 3.10 ≥ x > 3.14
* **vLLM**: 0.15.x (0.15.0 ≥ x > 0.16.0)

## Next up

Explore the Reka API's capabilities through the following guides:

* [Chatting with text, images, and video](/chat)
* [Visual understanding (from images to long-form videos)](/vision/overview)