DeepSeek V4-Pro

mit

DeepSeek-V4-Pro hybrid model with both non-thinking and thinking (default) modes.

Context
1M
Max output
384K
Input price
$0.44/1M tokens
Output price
$0.87/1M tokens
Snapshotsdeepseek-v4-pro-2026-04-24

Capabilities

vision
tool call
structured output
reasoning
json mode
streaming
fine tuning
batch

Details

Model IDdeepseek-v4-pro
Provider DeepSeek
Creator deepseek
Familydeepseek-chat
Licensemit
Parameters—
Statusactive
Input modalitiestext
Output modalitiestext
Architecture—
Knowledge cutoff—
Training data cutoff—
Release date
Deprecation date—
Typechat
Reasoning tokensYes
Max input—
Open weightYes
Sourceofficial
Last updated

Tools

Function Callingfunction_calling
Call external functions and APIs

Endpoints

Chat CompletionsPOST
Generate chat responses with messageshttps://api.deepseek.com/v1/chat/completions
CompletionsPOST
Legacy text completionhttps://api.deepseek.com/v1/completions

Pricing

Input
$0.44
Output
$0.87
Cache write
—
Cache read
$0
Batch in
—
Batch out
—

Family Comparison: deepseek-chat

ModelContextMax outPricing
DeepSeek V4-Flash1M384K$0.14$0.28
DeepSeek V3.2 (Chat)128K8K$0.28$0.42
DeepSeek V3.2 (Reasoner)128K64K$0.28$0.42
DeepSeek V4-Pro1M384K$0.44$0.87

API

GET/v1/models/deepseek/deepseek-v4-pro