Enter opens the highlighted result. Esc closes.

DeepSeek V4 Flash 0731 vs Devstral 2 2512

DeepSeek V4 Flash 0731 leads Devstral 2 2512 on the Artificial Analysis Intelligence Index, 34.3 to 8.6, and DeepSeek V4 Flash 0731 costs 5.1x less per output token ($0.396 vs $2.00 per million). Updated October 22, 2026; 5 benchmarks have a result for both.

Replace DeepSeek V4 Flash 0731 Replace Devstral 2 2512
DeepSeek V4 Flash 0731DeepSeek
Coding
51.8
Overall
55.2
Output $/M
$0.396
5
Devstral 2 2512Mistral AI
Coding
No data
Overall
No data
Output $/M
$2.00
0
vs boards won of 5 shared

DeepSeek V4 Flash 0731 or Devstral 2 2512: which to pick

Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; Devstral 2 2512 has none, so the two cannot be ranked on coding.

Compare them on price, context and the boards below.

Model ADeepSeek V4 Flash 0731DeepSeek51.8/1004 independent benchmarks$0.01 per solved taskModel BDevstral 2 2512Mistral AINot benchmarked1 independent benchmark$1.19 per solved task
Shared results
5
DeepSeek V4 Flash 0731 only
50
Devstral 2 2512 only
1
Like-for-like categories
3 of 9

Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.

Which one for your work

Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.

At a glance

Coding index
DeepSeek V4 Flash 0731 51.8 from 4 benchmarks, Devstral 2 2512 not benchmarked.
Widest category
Agentic work, 38.7 points apart on average over 1 shared benchmark.
Widest single gap
AA Agentic Index: DeepSeek V4 Flash 0731 41.0 against Devstral 2 2512 2.3, 38.7 points apart on a 0 to 100 scale.
Output price
DeepSeek V4 Flash 0731 $0.396 against Devstral 2 2512 $2.00 per million output tokens, 5.1x apart.
Context window
DeepSeek V4 Flash 0731 takes 1.05M tokens, Devstral 2 2512 262K.

Benchmark by benchmark

First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.

Boards won, where both have a result
CategoryDeepSeek V4 Flash 0731BarsDevstral 2 2512Ties
Coding30
Intelligence10
Agentic work10
All shared boards50
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
Coding benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
SWE-bench VerifiedNot measured53.8%Only Devstral 2 2512 measured
AA Coding Index69.131.3DeepSeek V4 Flash 0731
LiveCodeBench (Vals AI)87.3%51.8%DeepSeek V4 Flash 0731
SWE-bench Verified (Vals AI)88.8%62.8%DeepSeek V4 Flash 0731
Terminal-Bench 4 (Vals AI)18.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
Vibe Code Bench (Vals AI)74.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
Code migration (Vals AI)38.6%Not measuredOnly DeepSeek V4 Flash 0731 measured
ProgramBench (Vals AI)0.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
IOI (Vals AI)32.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
Terminal-Bench 2.1 (Vals AI)67.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
SkillsBench (Vals AI)50.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
CyberBench (Vals AI)72.1%Not measuredOnly DeepSeek V4 Flash 0731 measured
SciCode50.3%Not measuredOnly DeepSeek V4 Flash 0731 measured
FrontierCode18.8%Not measuredOnly DeepSeek V4 Flash 0731 measured
WeirdML63.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
ALE-Bench1306.1Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 23 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
AA Intelligence Index34.38.6DeepSeek V4 Flash 0731
Epoch Capabilities Index154.8Not measuredOnly DeepSeek V4 Flash 0731 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
Agentic work benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
AA Agentic Index41.02.3DeepSeek V4 Flash 0731
Finance Agent (Vals AI)49.5%Not measuredOnly DeepSeek V4 Flash 0731 measured
Tax Agent Bench (Vals AI)28.1%Not measuredOnly DeepSeek V4 Flash 0731 measured
Legal Research Bench (Vals AI)30.3%Not measuredOnly DeepSeek V4 Flash 0731 measured
Excel Modeling Benchmark (Vals AI)57.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
BioMysteryBench (Vals AI)64.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
Terminal-Bench-Science (Vals AI)4.3%Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 34 boards here
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
Reasoning benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
GPQA Diamond89.9%Not measuredOnly DeepSeek V4 Flash 0731 measured
GPQA Diamond (Artificial Analysis run)90.8%Not measuredOnly DeepSeek V4 Flash 0731 measured
Humanity's Last Exam (Artificial Analysis run)38.6%Not measuredOnly DeepSeek V4 Flash 0731 measured
GPQA Diamond (Epoch AI run)91.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
ARC-AGI89.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
ARC-AGI-261.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
SimpleBench61.1%Not measuredOnly DeepSeek V4 Flash 0731 measured
Chess puzzles32.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
Mystery game puzzles34.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 13 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Math benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
OTIS Mock AIME94.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
FrontierMath, tiers 1 to 357.5%Not measuredOnly DeepSeek V4 Flash 0731 measured
FrontierMath, tier 424.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
ProofBench56.0%Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 9 boards here
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Knowledge benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
MMLU Pro86.2%Not measuredOnly DeepSeek V4 Flash 0731 measured
LegalBench (Vals AI)77.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
CorpFin (Vals AI)61.9%Not measuredOnly DeepSeek V4 Flash 0731 measured
TaxEval (Vals AI)70.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
MedCode (Vals AI)41.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
SimpleQA Verified33.6%Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 12 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
Web design benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
Design Arena: websites1237Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: UI components1233Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: game dev1207Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: data viz1180Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: 3D1208Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: all code categories1229Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: SVG1184Not measuredOnly DeepSeek V4 Flash 0731 measured
Design Arena: ASCII art1090Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 1 board here
GeneralDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
General benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
MedScribe (Vals AI)80.4%Not measuredOnly DeepSeek V4 Flash 0731 measured
Long context reasoning (AA-LCR)79.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
DTBench90.9%Not measuredOnly DeepSeek V4 Flash 0731 measured
LMCA41.7%Not measuredOnly DeepSeek V4 Flash 0731 measured
Neither model has a result on the other 42 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for DeepSeek V4 Flash 0731 and Devstral 2 2512
BenchmarkDeepSeek V4 Flash 0731Devstral 2 2512Leader
Neither model has a result on 5 boards here

What each workload costs

Three fixed token mixes at each model's listed price, so per-token rates become a bill.

Chat turn

1K input + 500 output tokens

DeepSeek V4 Flash 0731$0.0003Fits in one request
Devstral 2 2512$0.0014Fits in one request

DeepSeek V4 Flash 0731 costs less

Repository review

50K input + 3K output tokens

DeepSeek V4 Flash 0731$0.0082Fits in one request
Devstral 2 2512$0.026Fits in one request

DeepSeek V4 Flash 0731 costs less

Agent loop

200K input + 10K output tokens

DeepSeek V4 Flash 0731$0.032Fits in one request
Devstral 2 2512$0.100Fits in one request

DeepSeek V4 Flash 0731 costs less

Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.

Specification differences

The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.

Specifications of DeepSeek V4 Flash 0731 and Devstral 2 2512
SpecificationDeepSeek V4 Flash 0731Mistral Devstral 2 2512
MakerDeepSeekMistral AI
WeightsOpen weightsOpen weights
ReleasedJuly 31, 2026December 9, 2025
Context window1.05M262K
Input $/Mtok$0.14$0.40
Output $/Mtok$0.396$2.00
Coding index51.8Not benchmarked
Independent measurements behind it41
Cost per solved task$0.01$1.19
SourceOpenRouter listingOpenRouter listing

Questions

Is DeepSeek V4 Flash 0731 better than Devstral 2 2512 for coding?

Only DeepSeek V4 Flash 0731 has a coding score, 51.8 on the Vibe Coding Index; Devstral 2 2512 has none, so the two cannot be ranked on coding. Compare them on price, context and the boards below.

Which is cheaper, DeepSeek V4 Flash 0731 or Devstral 2 2512?

DeepSeek V4 Flash 0731 costs $0.14 per million input tokens and $0.396 per million output tokens; Devstral 2 2512 costs $0.40 and $2.00. Per solved SWE-bench Verified task, DeepSeek V4 Flash 0731 costs $0.01 and Devstral 2 2512 $1.19.

Which has the bigger context window, DeepSeek V4 Flash 0731 or Devstral 2 2512?

DeepSeek V4 Flash 0731, with 1.05M tokens against 262K.

How many benchmarks is the DeepSeek V4 Flash 0731 vs Devstral 2 2512 comparison based on?

5 benchmarks have a result for both models as of October 22, 2026. DeepSeek V4 Flash 0731's coding index rests on 4 independent benchmarks and Devstral 2 2512's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next