Noometry

Model comparison

GPT-4.1 mini vs Step 3

Step 3 is the stronger model overall, scoring 40.5 to 33.6 on the Noometry Index.

Last verified . 17 shared benchmarks.

GPT-4.1 mini OpenAI

33.6

Rank #240 Confirmed

Step 3 StepFun

40.5

Rank #149 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GPT-4.1 mini scores higher in 2 categories and Step 3 in 7 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Step 3 leads 28.4 to 10.8.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 48.6% for GPT-4.1 mini and 62.3% for Step 3.
  • Step 3 has downloadable open weights; the other is API-only.

Side by side

GPT-4.1 miniStep 3
ProviderOpenAIStepFun
Noometry Index33.640.5
Released2025-04-14—
WeightsProprietaryOpen
Context window1.05M—
Max output33K—
Input $ / M tokens$0.40—
Output $ / M tokens$1.60—
Results tracked4717

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

Category by category

Coding Step 3 leads

GPT-4.1 mini: 30.6 (#293), Step 3: 40.1 (#147)

BenchmarkGPT-4.1 miniStep 3
LMArena Coding13671367
SWE-bench Verified (bash only)23.9%—
Aider Polyglot32.4%—
SciCode40.4%—
WeirdML37.6%—
BigCodeBench Instruct48.9%—
CadEval16%—

Agentic & Tool Use Not comparable

GPT-4.1 mini: 33.3 (#55), Step 3: —

BenchmarkGPT-4.1 miniStep 3
Berkeley Function Calling Leaderboard50.5%—

Reasoning Step 3 leads

GPT-4.1 mini: 10.8 (#340), Step 3: 28.4 (#105)

BenchmarkGPT-4.1 miniStep 3
Kagi LLM Benchmark48.6%62.3%
LMArena Hard Prompts13491355
ARC-AGI-20%—
ARC-AGI-13.5%—
CritPt0%—
Chess Puzzles7%—
Mystery Game Puzzles7%—
DTBench68.8%—
LMCA21.1%—
Epoch Capabilities Index135.01—

Math Step 3 leads

GPT-4.1 mini: 24.1 (#270), Step 3: 37.6 (#148)

BenchmarkGPT-4.1 miniStep 3
LMArena Math13431366
FrontierMath (Tiers 1-3)6.7%—
OTIS Mock AIME 2024-202544.7%—
Omni-MATH49.1%—
MATH Level 587.3%—
FrontierMath (Feb 2025 set)4.5%—

Knowledge Step 3 leads

GPT-4.1 mini: 34.7 (#194), Step 3: 36.8 (#164)

BenchmarkGPT-4.1 miniStep 3
LMArena Expert13381333
GPQA Diamond65.8%—
SimpleQA Verified12.7%—
MMLU-Pro78.3%—
GPQA (HELM)61.4%—

Multimodal Too close to call

GPT-4.1 mini: 35.8 (#82), Step 3: 35.5 (#86)

BenchmarkGPT-4.1 miniStep 3
LMArena Vision11811177

Multilingual Too close to call

GPT-4.1 mini: 45.7 (#166), Step 3: 46.3 (#159)

BenchmarkGPT-4.1 miniStep 3
LMArena Non-English13181327
LMArena Chinese13291397
LMArena German13511371
LMArena Korean12981269
LMArena Russian13241331
LMArena Spanish13191371
LMArena French1358—
LMArena Japanese1290—

Instruction Following GPT-4.1 mini leads

GPT-4.1 mini: 73.7 (#118), Step 3: 70.4 (#164)

BenchmarkGPT-4.1 miniStep 3
LMArena Instruction Following13331332
IFEval90.4%—

Long Context Step 3 leads

GPT-4.1 mini: 31.8 (#275), Step 3: 40.3 (#157)

BenchmarkGPT-4.1 miniStep 3
LMArena Longer Query13441326
Fiction.LiveBench44.4%—

Writing & Preference Step 3 leads

GPT-4.1 mini: 48.6 (#199), Step 3: 54.3 (#151)

BenchmarkGPT-4.1 miniStep 3
LMArena Text13401350
LMArena Creative Writing13001321
LMArena Multi-Turn13541341
EQ-Bench Creative Writing1147—
WildBench83.8%—

Frequently asked questions

Is GPT-4.1 mini better than Step 3?

Step 3 is the stronger model overall, scoring 40.5 to 33.6 on the Noometry Index.

Is GPT-4.1 mini or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 40.1 versus 30.6 in the Noometry coding category.

How many benchmarks do GPT-4.1 mini and Step 3 share?

17 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Step 3 has 17.

Related comparisons

Go deeper