# DeepSeek-V4.1-Flash

> DeepSeek's sparse mixture-of-experts model on the Causal Encoder-Decoder architecture, with reasoning and image input.

## Key facts

- Model author: DeepSeek
- Input: $0.2857 per 1M tokens (starting price)
- Output: $1.1429 per 1M tokens (starting price)
- Cached input: $0.005715 per 1M tokens
- Context window: 1,000,000 tokens
- Maximum output: 384,000 tokens
- Input modalities: image, text
- Output modalities: text
- Capabilities: reasoning, tools, vision
- Released: 2026-09-10
- Model IDs: `deepseek/deepseek-v4.1-flash`, `jd/deepseek-v4.1-flash`
- Model page: https://www.realrelay.ai/model/deepseek/deepseek-v4.1-flash

## Overview

DeepSeek-V4.1-Flash is a sparse mixture-of-experts model from DeepSeek and the first built on the company's Causal Encoder-Decoder (CED) architecture. It supports reasoning and tool calling, accepts text and image input, and returns text. Its large context window and output limit suit long inputs.

DeepSeek-V4.1-Flash has a 1M-token context window and can generate up to 384K output tokens. It accepts text and image input and returns text output. It was released on September 10, 2026.

## Capabilities

DeepSeek-V4.1-Flash supports reasoning, tool calling and image understanding. It accepts text and image as input and produces text as output.

Image input is part of the model, so DeepSeek-V4.1-Flash can read screenshots, charts and photographed documents alongside text. Reasoning and tool calling let it plan steps and call functions on the result.

## Pricing

DeepSeek-V4.1-Flash starts from $0.286 per 1M input tokens and from $1.14 per 1M output tokens. Cache reads start from $0.0057 per 1M tokens.

Pricing is tiered, and the price shown is the starting tier. DeepSeek-V4.1-Flash is available through RealRelay's OpenAI-compatible API.

## Use cases

Use DeepSeek-V4.1-Flash for document and screenshot understanding: it accepts text and image input, reasons over what it sees and returns a written answer, with up to 384K output tokens for long reports.

It also fits tool-using assistants and agents that must keep long histories in view, using the 1M-token context window.

## Call it

OpenAI-compatible chat completions:

```bash
curl https://www.realrelay.ai/v1/chat/completions \
  -H "Authorization: Bearer $REALRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek/deepseek-v4.1-flash", "messages": [{"role": "user", "content": "Hello"}]}'
```

## FAQ

### What is DeepSeek-V4.1-Flash?

DeepSeek-V4.1-Flash is a sparse mixture-of-experts model from DeepSeek and the first built on the company's Causal Encoder-Decoder (CED) architecture. It supports reasoning, tool calling and image understanding and accepts text and image input.

### How much does DeepSeek-V4.1-Flash cost?

DeepSeek-V4.1-Flash starts from $0.286 per 1M input tokens and from $1.14 per 1M output tokens, with cache reads from $0.0057 per 1M tokens. Pricing is tiered, and these are the starting-tier prices.

### What is the context window of DeepSeek-V4.1-Flash?

DeepSeek-V4.1-Flash has a 1M-token context window and can generate up to 384K tokens in a single response.

### What inputs and outputs does DeepSeek-V4.1-Flash support?

DeepSeek-V4.1-Flash accepts text and image input and returns text output.

### Does DeepSeek-V4.1-Flash support tool calling?

Yes. DeepSeek-V4.1-Flash supports tool calling, so it can call functions you define and use the results in multi-step tasks. It also supports reasoning and image understanding.

### When was DeepSeek-V4.1-Flash released?

DeepSeek-V4.1-Flash was released on September 10, 2026.

### How do I call DeepSeek-V4.1-Flash via the API?

Call DeepSeek-V4.1-Flash through RealRelay's OpenAI-compatible API using the model ID deepseek/deepseek-v4.1-flash.