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

# DeepSeek V4 Flash

> Call DeepSeek V4 Flash via Starrise AI OpenAI-compatible interface — fast, powerful, and easy to integrate.

DeepSeek V4 Flash is available through Starrise AI via the OpenAI-compatible interface. Supports thinking mode, function calling, and JSON mode.

## Key Capabilities

* **OpenAI-compatible** — Drop-in replacement for the OpenAI SDK, no other code changes needed
* **Thinking mode** — Deep thinking with configurable reasoning intensity
* **Function calling** — Native support for tools and function calls
* **JSON mode** — Structured output via `response_format`
* **Long context** — Supports large document processing and multi-turn conversations
* **Streaming** — Real-time token streaming via SSE
* **Prompt caching** — Automatic caching with hit/miss statistics returned

## Quick Example

<CodeGroup>
  ```bash cURL theme={null}
  curl https://ai.alad.com/v1/chat/completions \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model": "deepseek-v4-flash",
      "messages": [
        { "role": "user", "content": "Explain quantum entanglement in simple terms." }
      ]
    }'
  ```

  ```python Python theme={null}
  from openai import OpenAI

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

  response = client.chat.completions.create(
      model="deepseek-v4-flash",
      messages=[
          {"role": "user", "content": "Explain quantum entanglement in simple terms."}
      ]
  )

  print(response.choices[0].message.content)
  ```

  ```python Streaming theme={null}
  from openai import OpenAI

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

  stream = client.chat.completions.create(
      model="deepseek-v4-flash",
      messages=[
          {"role": "user", "content": "Write a short poem about the ocean."}
      ],
      stream=True
  )

  for chunk in stream:
      print(chunk.choices[0].delta.content or "", end="")
  ```
</CodeGroup>

## Parameters

| Parameter         | Type            | Required | Description                                                                                                                           |
| ----------------- | --------------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------- |
| `model`           | string          | Yes      | Fixed value: `deepseek-v4-flash`                                                                                                      |
| `messages`        | array           | Yes      | Array of `{ role, content }` objects; supports `system`, `user`, `assistant`, `tool` roles                                            |
| `thinking`        | object          | No       | Enable/disable thinking: `{"type": "enabled"}` or `{"type": "disabled"}`. Optional `reasoning_effort`: `high`, `max`, `low`, `medium` |
| `max_tokens`      | integer         | No       | Maximum number of tokens to generate                                                                                                  |
| `temperature`     | float           | No       | `0`–`2`, controls randomness, default `1`                                                                                             |
| `stream`          | boolean         | No       | Enable SSE streaming, default `false`                                                                                                 |
| `stream_options`  | object          | No       | `{"include_usage": true}` to return usage statistics at end of stream                                                                 |
| `top_p`           | float           | No       | Nucleus sampling threshold, default `1`                                                                                               |
| `stop`            | string / array  | No       | Sequences where generation stops (up to 16)                                                                                           |
| `response_format` | object          | No       | `{"type": "json_object"}` to enable JSON mode                                                                                         |
| `tools`           | array           | No       | List of function calling tools                                                                                                        |
| `tool_choice`     | string / object | No       | Tool selection control: `none`, `auto`, `required`, or specify a function                                                             |
| `user_id`         | string          | No       | Custom user ID for content safety and KVCache isolation                                                                               |

<Card title="API Reference" icon="code" href="/en/api-reference/model-api/deepseek/deepseek-v4-flash">
  View the interactive API Playground for DeepSeek V4 Flash.
</Card>