DeepSeek: DeepSeek V4 Pro

mit

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.

Context
1.0M
Max output
393K
Input price
$1.17/1M tokens
Output price
$2.34/1M tokens

Capabilities

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

Details

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

Tools

Function Callingfunction_calling
Call external functions and APIs

Pricing

Input
$1.17
↑169%$0.44
Output
$2.34
↑169%$0.87
Cache write
Cache read
$0.1
↑2375%$0
Batch in
Batch out
Price history · $/1M input
$1.74
$0.435
$1.168
Apr 25Apr 26Aug 13
DateInputOutputCache readChange
Apr 25, 2026$1.74$3.48$0.145
Apr 26, 2026da3af4a$0.435$0.87$0.004↓75%
Aug 13, 20263cddff5$1.168$2.336$0.099↑169%

Input is ↓33% since Apr 25, 2026, from $1.74 to $1.168 per 1M tokens.

Price across providers

ProviderInputOutputSince launch
Vercel AI Gatewaycheapest$0.43$0.87↓75%
OpenRouter$1.17$2.34↓33%
Novita AI$1.6$3.2↓5%

Family Comparison: deepseek

ModelContextMax outPricing
DeepSeek: DeepSeek V4 Flash (free)1.0M384K
DeepSeek: DeepSeek V4 Flash 04231.0M384K$0.06$0.12
DeepSeek V4 Flash Latest1.0M384K$0.07$0.13
Nex AGI: DeepSeek V3.1 Nex N1131K164K$0.14$0.5
DeepSeek: DeepSeek V4 Flash 07311.0M393K$0.14$0.28
DeepSeek: DeepSeek V3.2164K66K$0.27$0.4
DeepSeek: DeepSeek V3.1 Terminus164K33K$0.27$0.95
DeepSeek: DeepSeek V3.2 Exp164K66K$0.27$0.41
DeepSeek: DeepSeek V3.2 Speciale164K164K$0.29$0.43
DeepSeek: DeepSeek V4 Pro 08131.0M384K$0.44$0.87
DeepSeek: DeepSeek V4 Pro1.0M393K$1.17$2.34

API

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

DeepSeek: DeepSeek V4 Pro

deepseek/deepseek-v4-pro

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.

Changes · 3 entries
deepseek/deepseek-v4-proupdate3cddff5Aug 13, 2026, 02:54 AM
max_output_tokens384000393216
pricinginput$0.435$1.168output$0.87$2.336cached input$0.004$0.099
deepseek/deepseek-v4-proupdateda3af4aApr 26, 2026, 07:37 PM
pricinginput$1.74$0.435output$3.48$0.87cached input$0.145$0.004
deepseek/deepseek-v4-procreatec8065d6Apr 25, 2026, 10:39 AM