DeepSeek V4 Flash

DeepSeek · Efficient Models

#170.9

DeepSeek V4 Flash is DeepSeek's reasoning-capable model: $0.14/$0.28 per 1M input/output tokens, 1M token context window. It supports native tool calling, extended reasoning, open weights. Released 2026-04-24.

Verified · 2+ sourcesVerified Jun 10, 2026Long ContextOpen Weights

Input Price

$0.14

Output Price

$0.28

Cache Read Price

$0

Context Window

1M

All metrics

MetricValueProvenance receipt
Input Price$0.14
Output Price$0.28
Cache Read Price$0
Context Window1M
Max Output Tokens384K
Tool CallingYes
Extended ReasoningYes

Pricing

Pay-as-you-go API

$0.14per 1M input tokens

Output
$0.28 / 1M tokens
Cache read
$0 / 1M tokens
Single sourceChecked Jun 10, 2026
Official pricing →

Data sources

Facts on this page are sourced from the following verified sources.

All sources are verified before data goes live.

Compare DeepSeek V4 Flash

Related models

#7Gemini 3.5 Flash

Google

33.1

Gemini 3.5 Flash is Google's reasoning-capable model: $1.50/$9 per 1M input/output tokens, 1.0M token context window.

#9Claude Haiku 4.5 (latest)

Anthropic

32.8

Claude Haiku 4.5 (latest) is Anthropic's reasoning-capable model: $1/$5 per 1M input/output tokens, 200K token context window, 39.45% on SWE-Bench Pro.

#16Llama 4 Maverick 17B Instruct

Meta

21.1

Llama 4 Maverick 17B Instruct is Meta's general-purpose model: 1M token context window, 5.24% on SWE-Bench Pro.

#17o4-mini

OpenAI

20.4

o4-mini is OpenAI's reasoning-capable model: $1.10/$4.40 per 1M input/output tokens, 200K token context window.

Spotted an error? Suggest a correction.