DeepSeek-V4-Pro

mit

DeepSeek-V4-Pro-0813 hybrid model with both non-thinking and thinking (default) modes.

Context
1M
Max output
384K
Input price
$0.66/1M tokens
Output price
$1.98/1M tokens

Capabilities

Reasoning & Interaction
Reasoning
Tool Calling
Output Control
JSON Mode
Infrastructure
Streaming
Fine-tuning

Details

Provider Azure AI Foundry
Creator DeepSeek
Familydeepseek
Licensemit
Statusactive
Typechat
Input modalitiestext
Output modalitiestext
Open weightYes
Parameters—
Architecture—
Context window1M tokens
Max input—
Max output384K tokens
Reasoning tokensYes
Knowledge cutoff—
Release date
Sourceofficial

Tools

Function Callingfunction_calling
Call external functions and APIs

Pricing

Input
$0.66
Output
$1.98
Cache write
—
Cache read
$0.02
Batch in
—
Batch out
—

Family Comparison: deepseek

ModelContextMax out
DeepSeek-V3-0324——
DeepSeek-V3.1——
DeepSeek-V3.2-Speciale——
DeepSeek-V3.2——
DeepSeek-V4-Fast Preview——
DeepSeek-V4-Flash-0731 Preview——
DeepSeek-V4-Flash1M384K
DeepSeek-V4-Pro1M384K

API

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

DeepSeek-V4-Pro

DeepSeek-V4-Pro-0813 hybrid model with both non-thinking and thinking (default) modes.

Changes · 3 entries
deepseek-v4-procreateca7a400Nov 2, 2026, 03:02 AM
deepseek-v4-proupdate105275dJul 10, 2026, 12:00 AM
nameDeepSeek-V4-Pro Preview→DeepSeek-V4-Pro
deepseek-v4-procreate9e091d8May 23, 2026, 05:20 PM