Models/MiniMax-M2.1

MDL

MiniMax-M2.1

MiniMax

Verified against Artificial Analysis · Jul 13, 2026

Capability

percentile index

Price /M

$0.3 / $1.2

in / out

Context

204.8K

tokens

Avg score

64.0

10 benchmarks

AA Intelligence

31.4

index

Speed

79 tok/s

TTFT 1.2s

Released

Dec 23, 2025

About

MiniMax-M2.1 is a open-weights model in the MiniMax family from MiniMax. Benchmarked on 10 evals, averaging 64.0.

Benchmark scores · 10

avg 64.0 — higher is better

MMLU-Proartificial-analysis · 2026-07

87.5

τ²-benchartificial-analysis · 2026-07

85.4

GPQA Diamondartificial-analysis · 2026-07

AIMEartificial-analysis · 2026-07

82.7

LiveCodeBenchartificial-analysis · 2026-07

IFBenchartificial-analysis · 2026-07

69.9

AA Long-Context Reasoningartificial-analysis · 2026-07

SciCodeartificial-analysis · 2026-07

40.7

Terminal-Bench Hardartificial-analysis · 2026-07

28.8

Humanity's Last Examartificial-analysis · 2026-07

22.2

Details

PublisherMiniMax

FamilyMiniMax

Parameters—

License—

WeightsOpen

ReleasedDec 23, 2025

Performance & price

Output speed79 tok/s

Latency (TTFT)1.2s

First answer26s

Price in / out$0.3 / $1.2 per 1M

Blended price$0.525 / 1M

Math index82.7

Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →

History

full ledger →

Capability over time

Append-only ledger — every observed change to this model's numbers.

API providers

Artificial Analysis$1.2/M