Model comparison

MiniMax-M3 vs Step 2 16k Exp 202412

MiniMax-M3 is the stronger model overall, scoring 43.5 to 38.6 on the Noometry Index.

Last verified . 12 shared benchmarks.

MiniMax-M3 MiniMax

43.5

Rank #81 Confirmed

Step 2 16k Exp 202412 StepFun

38.6

Rank #173 Confirmed

Summary

  • They share 12 benchmarks with published results for both. MiniMax-M3 scores higher in 8 categories and Step 2 16k Exp 202412 in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 34.6.
  • MiniMax-M3 has downloadable open weights; the other is API-only.

Side by side

MiniMax-M3 and Step 2 16k Exp 202412 specifications
MiniMax-M3Step 2 16k Exp 202412
ProviderMiniMaxStepFun
Noometry Index43.538.6
Released2026-06-01—
WeightsOpenProprietary
Context window1M—
Max output512K—
Input $ / M tokens$0.30—
Output $ / M tokens$1.20—
Results tracked4112

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

Category by category

Coding MiniMax-M3 leads

MiniMax-M3: 41.5 (#111), Step 2 16k Exp 202412: 37.9 (#170)

Coding benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Coding14701316
FrontierCode14.7%—
LMArena WebDev1482—
SciCode47.1%—
ALE-Bench640.02—

Agentic & Tool Use Not comparable

MiniMax-M3: 22.5 (#121), Step 2 16k Exp 202412: —

Agentic & Tool Use benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
APEX-Agents37.7%—
OSWorld 2.04.6%—
GBAEval0.9%—
Vending-Bench 22,158—

Reasoning MiniMax-M3 leads

MiniMax-M3: 29.3 (#83), Step 2 16k Exp 202412: 24.8 (#140)

Reasoning benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Hard Prompts14471299
SimpleBench45.8%—
NYT Connections (extended)65.1%—
CritPt3.7%—
Chess Puzzles14%—
Mystery Game Puzzles8%—
DTBench78.9%—
LMCA33.7%—
Surface Evolver Bench55%—
Epoch Capabilities Index147—
ForecastBench61.4—

Math MiniMax-M3 leads

MiniMax-M3: 39.1 (#83), Step 2 16k Exp 202412: 33.6 (#171)

Math benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Math14321305
OTIS Mock AIME 2024-202571.1%—
ProofBench18%—

Knowledge MiniMax-M3 leads

MiniMax-M3: 58.4 (#32), Step 2 16k Exp 202412: 34.6 (#187)

Knowledge benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Expert14621279
GPQA Diamond90.9%—

Multimodal Not comparable

MiniMax-M3: 39.5 (#45), Step 2 16k Exp 202412: —

Multimodal benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Vision1254—
LMArena Document1435—

Multilingual MiniMax-M3 leads

MiniMax-M3: 53.2 (#70), Step 2 16k Exp 202412: 43.9 (#176)

Multilingual benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Non-English14201290
LMArena Chinese14651331
LMArena Russian14301324
LMArena French1445—
LMArena German1424—
LMArena Japanese1381—
LMArena Korean1372—
LMArena Spanish1431—

Instruction Following MiniMax-M3 leads

MiniMax-M3: 75.7 (#56), Step 2 16k Exp 202412: 68.1 (#187)

Instruction Following benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Instruction Following14331287

Long Context MiniMax-M3 leads

MiniMax-M3: 44.2 (#68), Step 2 16k Exp 202412: 39.7 (#165)

Long Context benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Longer Query14451307

Writing & Preference MiniMax-M3 leads

MiniMax-M3: 62.3 (#79), Step 2 16k Exp 202412: 52.5 (#168)

Writing & Preference benchmarks
BenchmarkMiniMax-M3Step 2 16k Exp 202412
LMArena Text14331321
LMArena Creative Writing14051328
LMArena Multi-Turn14411292
EQ-Bench 41150—

Frequently asked questions

Is MiniMax-M3 better than Step 2 16k Exp 202412?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 38.6 on the Noometry Index.

Is MiniMax-M3 or Step 2 16k Exp 202412 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 37.9 in the Noometry coding category.

How many benchmarks do MiniMax-M3 and Step 2 16k Exp 202412 share?

12 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Step 2 16k Exp 202412 has 12.

Related comparisons

Go deeper