Noometry

Model comparison

MiniMax-M2.1 vs MiniMax-M3

MiniMax-M3 is the stronger model overall, scoring 43.8 to 38.9 on the Noometry Index.

Last verified . 20 shared benchmarks.

MiniMax-M2.1 MiniMax

38.9

Rank #178 Confirmed

MiniMax-M3 MiniMax

43.8

Rank #85 Confirmed

Summary

  • They share 20 benchmarks with published results for both. MiniMax-M2.1 scores higher in 1 category and MiniMax-M3 in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 38.3.
  • The biggest single-benchmark swing is NYT Connections (extended): 11.2% for MiniMax-M2.1 and 65.1% for MiniMax-M3.
  • Both cost about the same: $0.30 input and $1.20 output per million tokens.
  • MiniMax-M3 accepts more context: 1M tokens versus 205K.

Side by side

MiniMax-M2.1MiniMax-M3
ProviderMiniMaxMiniMax
Noometry Index38.943.8
Released2025-12-232026-06-01
WeightsOpenOpen
Context window205K1M
Max output131K512K
Input $ / M tokens$0.30$0.30
Output $ / M tokens$1.20$1.20
Results tracked2241

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

Category by category

Coding MiniMax-M3 leads

MiniMax-M2.1: 40.4 (#143), MiniMax-M3: 41.8 (#118)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena WebDev13841482
LMArena Coding14211469
ALE-Bench623.83640.02
FrontierCode—14.7%
SciCode—47.1%

Agentic & Tool Use MiniMax-M2.1 leads

MiniMax-M2.1: 27.9 (#98), MiniMax-M3: 22.6 (#130)

BenchmarkMiniMax-M2.1MiniMax-M3
Terminal-Bench36.6%—
APEX-Agents—37.7%
OSWorld 2.0—4.6%
GBAEval—0.9%
Vending-Bench 2—2,158

Reasoning MiniMax-M3 leads

MiniMax-M2.1: 16.6 (#302), MiniMax-M3: 30.1 (#87)

BenchmarkMiniMax-M2.1MiniMax-M3
NYT Connections (extended)11.2%65.1%
LMArena Hard Prompts14111447
SimpleBench—45.8%
CritPt—3.7%
Chess Puzzles—14%
Mystery Game Puzzles—8%
DTBench—78.9%
LMCA—33.7%
Surface Evolver Bench—55%
Epoch Capabilities Index—146.95
ForecastBench—61.4

Math MiniMax-M3 leads

MiniMax-M2.1: 38.3 (#138), MiniMax-M3: 40.0 (#95)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Math13971429
OTIS Mock AIME 2024-2025—71.1%
ProofBench—18%

Knowledge MiniMax-M3 leads

MiniMax-M2.1: 38.3 (#147), MiniMax-M3: 58.4 (#35)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Expert14311461
GPQA Diamond—90.9%
Vectara Hallucination Rate11.8%—

Multimodal Not comparable

MiniMax-M2.1: —, MiniMax-M3: 40.2 (#51)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Vision—1253
LMArena Document—1435

Multilingual MiniMax-M3 leads

MiniMax-M2.1: 50.0 (#128), MiniMax-M3: 53.0 (#75)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Non-English13781420
LMArena Chinese14301463
LMArena French14041447
LMArena German13811426
LMArena Japanese12871381
LMArena Korean12981372
LMArena Russian13871428
LMArena Spanish13971432

Instruction Following MiniMax-M3 leads

MiniMax-M2.1: 73.8 (#112), MiniMax-M3: 75.5 (#62)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Instruction Following14001433

Long Context Too close to call

MiniMax-M2.1: 43.2 (#101), MiniMax-M3: 44.2 (#72)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Longer Query14161445

Writing & Preference MiniMax-M3 leads

MiniMax-M2.1: 58.3 (#120), MiniMax-M3: 62.1 (#83)

BenchmarkMiniMax-M2.1MiniMax-M3
LMArena Text13921433
LMArena Creative Writing13611404
LMArena Multi-Turn13961442
EQ-Bench 4—1150

Frequently asked questions

Is MiniMax-M2.1 better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.8 to 38.9 on the Noometry Index.

Which is cheaper, MiniMax-M2.1 or MiniMax-M3?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; MiniMax-M2.1 lists at $0.30 and $1.20.

Is MiniMax-M2.1 or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 40.4 in the Noometry coding category.

Which has the bigger context window?

MiniMax-M3 does, with 1M tokens against 205K.

How many benchmarks do MiniMax-M2.1 and MiniMax-M3 share?

20 benchmarks have published results for both models. MiniMax-M2.1 has 22 scored results on Noometry and MiniMax-M3 has 41.

Related comparisons

Go deeper