# DeepSeek V4.1 Flash

## Best for

Fast coding agents that need image input and a long context

## When not to use

Audio or video input. Not V4 Pro — keep Pro on the main agent and this Flash on sub-agents. Not V4 Flash 0731 (`dsv4`) when you want the cheaper text-only path.

## Overview

DeepSeek V4.1 Flash is a 552B-parameter mixture-of-experts model. It activates 8B parameters for input tokens and 16B for output tokens. The model handles coding, reasoning, images, and tool use, with a 1,048,576-token context window. Its smaller KV cache can also reduce costs for long-running requests that reuse context.

| | |
| --- | --- |
| API identifier | `deepseek/deepseek-v4.1-flash` |
| Developer | DeepSeek |
| Context window | 1,048,576 tokens |
| Supports | text, reasoning, tool-use, image input, vision |
| Status | Serving |

## Rates

USD per 1M tokens, on-demand tier.

| Input | Cached input | Output |
| --- | --- | --- |
| $0.20 | $0.01 | $1.00 |

## Call it

Put `deepseek/deepseek-v4.1-flash` in the `model` field. The same request works against the Anthropic Messages API — see [API Compatibility](https://inference.boundless.network/docs/api-compatibility).

```bash
curl https://api.inference.boundless.network/v1/chat/completions \
  -H "Authorization: Bearer $BOUNDLESS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4.1-flash",
    "messages": [{
      "content": "Say hello",
      "role": "user"
    }]
  }'
```

Every model we serve is listed at [Models](https://inference.boundless.network/models); the rate card in full is on [Pricing](https://inference.boundless.network/docs/pricing).