NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/DeepSeek V4 Flash 0731
DeepSeek logo

DeepSeek V4 Flash 0731

by DeepSeek · Released Jul 2026

1M Context
58.9
avg score
Rank #97
Compare
Better than 70% of all models
Context
1.0M tokens (~524 books)
Input $/1M
$0.04
Output $/1M
$0.30
Type
text
License
Proprietary
Benchmarks
15 tested
Data as of
About

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....

Tested on 15 benchmarks · BenchGecko score 58.9. Top scores: OTIS Mock AIME 2024-2025 (94.4%), GPQA diamond (91.7%), ARC-AGI (89.0%).

Capabilities
coding
40.9
#136 globally
reasoning
67.9
#58 globally
math
59.8
#102 globally
knowledge
51.3
#127 globally
general
54.3
#30 globally
Benchmark Scores
Compare All
Tested on 15 benchmarks · Ranked across 5 categories
Score Distribution (all 326 models)
0255075100
▲ You are here
codingCompare coding →
WeirdML

Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.

63.0·
Frontiercode
18.8·
reasoningCompare reasoning →
ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

89.0·
ARC-AGI-2

ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.

61.4·
SimpleBench

Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.

53.3·
mathCompare math →
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

94.4·
FrontierMath-Tiers-1-3-v2-Private
60.6·
FrontierMath-Tier-4-v2-Private
24.4·
Quick compare:vs Nemotron-4 15BDeepSeek logovs DeepSeek R1 Distill Qwen 32BGoogle DeepMind logovs Gemini 3.6 Flash
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · DeepSeek DeepSeek V4
Apr 2026
68.6
$0.09/M in1.0Mctx16 benchmarks
Jul 2026
58.9-9.7
$0.04/M in(-0.05)1.0Mctx15 benchmarks
Jul 2026
$0.05/M in(+0.00)1.0Mctx
Jul 2026
$0.00/M in(-0.05)1.0Mctx
Aug 2026
$0.44/M in(+0.44)1.0Mctx
Aug 2026
$0.11/M in(-0.33)1.0Mctx
Apr 2026
60.5+1.6
$0.88/M in(+0.77)1.0Mctx34 benchmarks
Aug 2026
62.6+2.1
$0.11/M in(-0.77)1.0Mctx22 benchmarks
Aug 2026
$0.66/M in(+0.55)1.0Mctx
See the full DeepSeek V4 family →
Recently Happened
DeepSeek V4 Flash 0731 pricing dropped 26%
Nov 8, 2026
DeepSeek V4 Flash 0731 pricing dropped 50%
Nov 4, 2026
DeepSeek V4 Flash 0731 pricing dropped 82%
Oct 29, 2026
DeepSeek V4 Flash 0731 pricing increased 780%
Oct 20, 2026
DeepSeek V4 Flash 0731 pricing increased 132%
Oct 15, 2026
DeepSeek V4 Flash 0731 pricing increased 198%
Oct 4, 2026
Similar Models
Nemotron-4 15B
Unknown
58.9n/a
DeepSeek logo
DeepSeek R1 Distill Qwen 32B
DeepSeek
58.7n/a
Google DeepMind logo
Gemini 3.6 Flash
Google DeepMind
58.6$0.75/1M
Links
Info
DeepSeekPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@DeepSeek
BenchGecko API
deepseek-v4-flash-0731
Specifications
  • Typetext
  • Context1.0M tokens (~524 books)
  • ReleasedJul 2026
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.000
Available On
DeepSeek logoDeepSeek$0.04
Categories
codingreasoningmathknowledgegeneral
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
Nemotron-4 15B
58.9
DeepSeek logoDeepSeek R1 Distill Qwen 32B
58.7
Google DeepMind logoGemini 3.6 Flash
58.6
OpenAI logoGPT-5
58.5
DeepSeek V4 Flash 0731 is a proprietary text AI model by DeepSeek, released in July 2026. It has an average benchmark score of 58.9. Context window: 1M tokens.

Related Models

Nemotron-4 15B · UnknownDeepSeek R1 Distill Qwen 32B · DeepSeekGemini 3.6 Flash · Google DeepMindGPT-5 · OpenAIGemini 2.5 Pro · Google DeepMind

Benchmarks

OTIS Mock AIME 2024-2025GPQA diamondARC-AGIDtbenchWeirdML

Related Pages

DeepSeek · ProviderDeepSeek · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-11-11

  • DeepSeek V4 Flash 0731 by DeepSeek. BenchGecko score 58.9, rank 96 of 326 scored models (normalized average of public benchmark scores).
  • List price $0.0434 input · $0.30 output per 1M tokens (as of 2026-11-11).
  • Sold by 27 providers (as of 2026-11-11): Wafer (nvfp4) $0.0434 in / $0.30 out · DeepInfra (fp8) $0.0600 in / $0.18 out · Relace (fp4) $0.0700 in / $0.14 out · Reka $0.10 in / $0.28 out · DigitalOcean $0.12 in / $0.24 out · and 22 more. Every provider
  • Gecko Tests · Gecko Score 26 (rank 49): Who Are You E (Sometimes says it is Google or Anthropic) · Censorship Index D (Refuses or dodges 5%) · Knowledge Horizon D (Knows news up to Jun 2025) · Tokenizer Tax D (68% more tokens outside English).

How to cite · data as of 2026-11-11

DeepSeek V4 Flash 0731 · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-11. https://benchgecko.ai/model/deepseek-v4-flash-0731

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP