DeepSeek V4-Pro-0813

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
Snapshotsdeepseek-v4-pro-2026-08-13

Capabilities

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

Details

Model IDdeepseek-v4-pro
Provider DeepSeek
Creator deepseek
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

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

Family Comparison: deepseek-chat

ModelContextMax outPricing
DeepSeek V4———
DeepSeek V3.2 (Chat)—384K$0.14$0.28
DeepSeek V3.2 (Chat)—384K$0.14$0.28
DeepSeek V4-Flash1M384K$0.14$0.28
DeepSeek V4.1-Flash-10171M384K$0.15$0.6
DeepSeek V3.2 (Chat)128K8K$0.28$0.42
DeepSeek V3.2 (Reasoner)128K64K$0.28$0.42
DeepSeek V41M384K$0.66$1.98
DeepSeek V4-Pro-08131M384K$0.66$1.98

API

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

DeepSeek V4-Pro-0813

deepseek-v4-pro

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

Changes · 16 entries
deepseek-v4-proupdateca7a400Nov 2, 2026, 03:02 AM
nameDeepSeek V3.2 (Chat)→DeepSeek V4-Pro-0813
descriptionDeepSeek-V3.2 in non-thinking mode. Best for general chat...→DeepSeek-V4-Pro-0813 hybrid model with both non-thinking ...
taglineDeepSeek-V3.2 in non-thinking mode.→DeepSeek-V4-Pro hybrid model with both non-thinking and t...
capabilitiesreasoningyes
pricinginput$0.14→$0.66output$0.28→$1.98cached input$0.0028→$0.022
endpoints["chat_completions"]→["chat_completions","completions"]
release_date2026-08-13
context_window1000000
max_output_tokens384000
reasoning_tokensyes
snapshots["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31...
deepseek-v4-proupdateb2f45bfOct 29, 2026, 03:29 AM
nameDeepSeek V4→DeepSeek V4-Pro-0813
descriptionDeepSeek-V4 hybrid model with both non-thinking and think...→DeepSeek-V4-Pro-0813 hybrid model with both non-thinking ...
pricinginput$0.15→$0.66output$0.6→$1.98cached input$0.003→$0.022
deepseek-v4-proupdate4e8b6afOct 21, 2026, 06:35 AM
nameDeepSeek V4-Pro→DeepSeek V4
descriptionDeepSeek-V4-Pro hybrid model with both non-thinking and t...→DeepSeek-V4 hybrid model with both non-thinking and think...
deepseek-v4-proupdate70e2981Sep 24, 2026, 11:29 AM
nameDeepSeek V4→DeepSeek V4-Pro
descriptionDeepSeek-V4 hybrid model with both non-thinking and think...→DeepSeek-V4-Pro hybrid model with both non-thinking and t...
deepseek-v4-proupdate118451aSep 24, 2026, 07:08 AM
nameDeepSeek V4-Pro→DeepSeek V4
descriptionDeepSeek-V4-Pro hybrid model with both non-thinking and t...→DeepSeek-V4 hybrid model with both non-thinking and think...
deepseek-v4-proupdate265df08Sep 12, 2026, 06:33 AM
release_date2026-09-10→2026-08-13
deepseek-v4-proupdate8e0a0b7Sep 10, 2026, 06:38 AM
statusdeprecated→active
release_date2026-08-13→2026-09-10
pricinginput$0.22→$0.15output$0.66→$0.6cached input$0.007→$0.003
snapshots["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31...→["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31...
deepseek-v4-proupdate5cd8ca9Sep 6, 2026, 06:34 AM
statusactive→deprecated
deepseek-v4-proupdatee3d7dedAug 31, 2026, 04:13 AM
pricinginput$0.435→$0.22output$0.87→$0.66cached input$0.003625→$0.007
deepseek-v4-proupdatebe35184Aug 17, 2026, 04:02 AM
release_date2026-07-31→2026-08-13
snapshots["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31"]→["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31...
deepseek-v4-proupdate39b8745Jul 31, 2026, 06:11 AM
release_date2026-04-24→2026-07-31
snapshots["deepseek-v4-pro-2026-04-24"]→["deepseek-v4-pro-2026-04-24","deepseek-v4-pro-2026-07-31"]
deepseek-v4-proupdatebf78dcfJul 25, 2026, 05:49 AM
nameDeepSeek V4→DeepSeek V4-Pro
descriptionDeepSeek-V4 hybrid model with both non-thinking and think...→DeepSeek-V4-Pro hybrid model with both non-thinking and t...
deepseek-v4-proupdate20c73acJul 17, 2026, 02:34 AM
nameDeepSeek V4-Pro→DeepSeek V4
descriptionDeepSeek-V4-Pro hybrid model with both non-thinking and t...→DeepSeek-V4 hybrid model with both non-thinking and think...
deepseek-v4-proupdate08f0ffeMay 11, 2026, 08:27 AM
nameDeepSeek V3.2 (Chat)→DeepSeek V4-Pro
descriptionDeepSeek-V3.2 in non-thinking mode. Best for general chat...→DeepSeek-V4-Pro hybrid model with both non-thinking and t...
taglineDeepSeek-V3.2 in non-thinking mode.→DeepSeek-V4-Pro hybrid model with both non-thinking and t...
pricinginput$0.14→$0.435output$0.28→$0.87cached input$0.0028→$0.003625
endpoints["chat_completions"]→["chat_completions","completions"]
release_date2026-04-24
context_window1000000
max_output_tokens384000
reasoning_tokensyes
snapshots["deepseek-v4-pro-2026-04-24"]
deepseek-v4-proupdateda3af4aApr 26, 2026, 07:37 PM
pricingcached input$0.028→$0.0028
deepseek-v4-procreatec8065d6Apr 25, 2026, 10:39 AM