Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Balance
Subscription
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksEarn crypto
  1. Models
  2. Text
  3. DeepSeek

DeepSeek V4 Pro

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with a 1M-token context window, built for advanced reasoning, coding, long-horizon agent workflows, knowledge, math, and software engineering tasks.

  • Reasoning
  • Tool Calling
  • Structured Output

Added Apr 24, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.70
Output
$1.40
Cache read
$0.07
Compare provider prices

Specifications

Context window
1M
Max output
384K
Parameters
1.6T / 49B
Total / active
Avg output (7d)
480 tokens
Longer than 41% of models

Benchmarks

Sourced from LMArena.

Arena Score

1457.9

Overall Rank

#64 / 419

Votes

57,481

Confidence Interval

1454.0 - 1461.8

Category Scores

Coding

#80 / 414

17,110 votes

1500.4

Math

#88 / 404

3,059 votes

1443.8

Longer Query

#61 / 397

25,876 votes

1471.7

Creative Writing

#46 / 417

9,691 votes

1446.4

Instruction Following

#61 / 419

19,975 votes

1452.8

Hard Prompts

#66 / 419

38,433 votes

1480.0

Additional Categories
22

Japanese

#41 / 288

720 votes

1454.3

French

#45 / 301

2,210 votes

1486.5

German

#48 / 320

996 votes

1468.3

Industry Legal And Government

#51 / 392

4,486 votes

1474.4

Multi Turn

#51 / 417

10,350 votes

1475.2

Industry Writing And Literature And Language

#55 / 418

14,095 votes

1451.7

Polish

#56 / 244

1,124 votes

1464.1

Industry Life And Physical And Social Science

#58 / 417

9,155 votes

1481.2

Industry Business And Management And Financial Operations

#60 / 412

11,358 votes

1459.6

Industry Entertainment And Sports And Media

#61 / 417

12,771 votes

1435.4

Russian

#61 / 383

6,300 votes

1460.6

Spanish

#63 / 302

1,727 votes

1456.4

Exclude Ties

#64 / 419

43,532 votes

1461.8

Industry Medicine And Healthcare

#64 / 390

4,052 votes

1477.8

English

#66 / 419

25,239 votes

1467.3

Non English

#66 / 419

32,242 votes

1444.9

Hard Prompts English

#67 / 417

17,810 votes

1484.8

Korean

#68 / 290

1,112 votes

1416.6

Industry Software And It Services

#71 / 419

23,616 votes

1491.5

Expert

#76 / 369

5,896 votes

1479.1

Industry Mathematical

#76 / 402

3,288 votes

1459.7

Chinese

#80 / 395

3,383 votes

1488.5

Published 2026-11-06 · Matched as DeepSeek-V4-Pro

LMArena Dataset

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Auto
Auto (ZDR)
$1.10$2.20$0.11N/AN/A
Auto
Auto (Anonymized)
$0.70$1.40$0.07N/AN/A
Relace
Relace
FP4
UNK
$0.735$1.47$0.07350.9s89 tps
Runware
Runware
10% off
UNK
$1.01$0.908$2.02$1.82$0.0829$0.0747N/AN/A
Uomi
Uomi
BF16
GLB
$1.22$2.39N/AN/AN/A
DeepInfra
DeepInfra
FP8
US
$1.36$2.73$0.1051.3s39 tps
Ollama
Ollama
UNK
$1.39$4.16$0.0462N/AN/A
Venice
Venice
US
$1.73$3.47$0.3472s34 tps
AtlasCloud
AtlasCloud
FP4
US
$1.76$3.55$0.1361.7s32 tps
RunAnywhere Fast (Wally)
RunAnywhere Fast (Wally)
FP8
US
$1.76$3.55$0.147N/AN/A
SiliconFlow
SiliconFlow
FP8
SG
$1.83$3.65$0.1521.1s40 tps
StreamLake
StreamLake
FP8
CN
$1.83$3.65$0.1522.6s38 tps
Nebius
Nebius
FP8
GLB
$1.84$3.68$0.1470.7s46 tps
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclearBest latency and throughput are highlighted.
Batch API
ProviderInputOutputCached input
Auto
Autodeepseek/deepseek-v4-pro
$0.70/M$1.40/M$0.07/M

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Related text models

Compare DeepSeek V4 Pro with similar models from the same provider or model family.

DeepSeek V4 Pro 0813

deepseek/deepseek-v4-pro-0813

DeepSeek V4 Pro 0813 is the general-availability release of DeepSeek V4 Pro, built for coding, tool use, cybersecurity, automation, and long-horizon agent workflows. DeepSeek reports strong gains over the preview and scores above Opus 4.8 on Terminal Bench 2.1, Cybergym, DeepSWE, and AutomationBench. It supports a 1M-token context window.

DeepSeek V4 Pro 0813 Thinking

deepseek/deepseek-v4-pro-0813:thinking

DeepSeek V4 Pro 0813 Thinking enables reasoning by default on the general-availability release of DeepSeek V4 Pro, built for coding, tool use, cybersecurity, automation, and long-horizon agent workflows. DeepSeek reports strong gains over the preview and scores above Opus 4.8 on Terminal Bench 2.1, Cybergym, DeepSWE, and AutomationBench. It supports a 1M-token context window.

DeepSeek V4 Pro (Thinking)

deepseek/deepseek-v4-pro:thinking

DeepSeek V4 Pro Thinking enables DeepSeek's chain-of-thought mode on the large-scale Mixture-of-Experts model with a 1M-token context window, built for advanced reasoning, coding, long-horizon agent workflows, knowledge, math, and software engineering tasks.

DeepSeek V4.1 Flash 0910 Uncensored

deepseek/deepseek-v4.1-flash-0910-uncensored

Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility; the served checkpoint can vary with availability.

DeepSeek V4.1 Flash 0910 Uncensored Thinking

deepseek/deepseek-v4.1-flash-0910-uncensored:thinking

Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility; the served checkpoint can vary with availability.

DeepSeek V4.1 Flash 1017 Uncensored

deepseek/deepseek-v4.1-flash-1017-uncensored

Uncensored DeepSeek V4.1 Flash with text and image input, optional reasoning, and tool calling. The date label is retained for compatibility; the served checkpoint can vary with availability.