DeepSeek V4.1 Flash 1017

• Type: Generation
• Capabilities: vision, reasoning
• Cache read: $0.01 per 1M input tokens (0.02× standard input price). See prompt caching.

Overview

DeepSeek V4.1 Flash 1017 is the October 17, 2026 update for coding, reasoning, tool use, and long-context analysis. It accepts text and image inputs and generates text, with tool calling, structured outputs, and a context window of up to 1,048,576 tokens across input and output. Set reasoning_effort to none to disable thinking, or low, high, or max to select a reasoning level. minimal maps to low, medium to high, and xhigh to max.

Reasoning efforts

• Supported: none, minimal, low, medium, high, xhigh, max

See the reasoning effort guide for request examples.

Pricing

Priority	Input Tokens (per 1M)	Cache Read Tokens (per 1M)	Output Tokens (per 1M)
Realtime1	$0.15	$0.01	$0.60
Async	$0.12	$0.01	$0.48
Batch (24h)	$0.08	$0.01	$0.30

Playground

Open this model in the Playground.

Footnotes

1. Realtime availability is limited. Doubleword is primarily a batch API. ↩