Model comparison

DeepSeek V4.1 Flash vs Step 3

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.5 on the Noometry Index.

Last verified . 16 shared benchmarks.

DeepSeek V4.1 Flash DeepSeek

52.8

Rank #38 Confirmed

Step 3 StepFun

40.5

Rank #149 Confirmed

Summary

  • They share 16 benchmarks with published results for both. DeepSeek V4.1 Flash scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where DeepSeek V4.1 Flash leads 66.7 to 37.6.

Side by side

DeepSeek V4.1 Flash and Step 3 specifications
DeepSeek V4.1 FlashStep 3
ProviderDeepSeekStepFun
Noometry Index52.840.5
Released2026-09-09—
WeightsOpenOpen
Context window1M—
Max output393K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked3717

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 52.9 (#32), Step 3: 40.1 (#147)

Coding benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Coding15061367
LMArena WebDev1619—
SciCode51.9%—
ALE-Bench1,092—

Agentic & Tool Use Not comparable

DeepSeek V4.1 Flash: 31.2 (#69), Step 3: —

Agentic & Tool Use benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
APEX-Agents39.5%—
GDP.pdf19.8%—

Reasoning DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 50.2 (#36), Step 3: 28.4 (#105)

Reasoning benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Hard Prompts14831355
Kagi LLM Benchmark—62.3%
NYT Connections (extended)89.6%—
CritPt14.3%—
Mystery Game Puzzles43%—
DTBench89.9%—
LMCA47%—
Surface Evolver Bench46.3%—
Epoch Capabilities Index154.9—

Math DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 66.7 (#25), Step 3: 37.6 (#148)

Math benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Math14771366
FrontierMath (Tiers 1-3)67.4%—
FrontierMath Tier 426.8%—
OTIS Mock AIME 2024-202598.3%—
ProofBench54%—

Knowledge DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 57.9 (#38), Step 3: 36.8 (#164)

Knowledge benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Expert15061333
GPQA Diamond89.8%—

Multimodal DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 39.1 (#61), Step 3: 35.5 (#86)

Multimodal benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Vision12771177
Furniture Assembly34.2%—

Multilingual DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 55.0 (#35), Step 3: 46.3 (#159)

Multilingual benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Non-English14481327
LMArena Chinese14971397
LMArena German14841371
LMArena Korean14521269
LMArena Russian14711331
LMArena Spanish14591371
LMArena French1452—
LMArena Japanese1412—

Instruction Following DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 77.3 (#26), Step 3: 70.4 (#164)

Instruction Following benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Instruction Following14741332

Long Context DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 45.2 (#47), Step 3: 40.3 (#157)

Long Context benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Longer Query14751326

Writing & Preference DeepSeek V4.1 Flash leads

DeepSeek V4.1 Flash: 65.4 (#48), Step 3: 54.3 (#151)

Writing & Preference benchmarks
BenchmarkDeepSeek V4.1 FlashStep 3
LMArena Text14621350
LMArena Creative Writing14351321
LMArena Multi-Turn14571341
EQ-Bench Creative Writing1540—

Frequently asked questions

Is DeepSeek V4.1 Flash better than Step 3?

DeepSeek V4.1 Flash is the stronger model overall, scoring 52.8 to 40.5 on the Noometry Index.

Is DeepSeek V4.1 Flash or Step 3 better for coding?

DeepSeek V4.1 Flash scores higher on coding benchmarks: 52.9 versus 40.1 in the Noometry coding category.

How many benchmarks do DeepSeek V4.1 Flash and Step 3 share?

16 benchmarks have published results for both models. DeepSeek V4.1 Flash has 37 scored results on Noometry and Step 3 has 17.

Related comparisons

Go deeper