Noometry

Model comparison

GPT-4 vs MiniMax-M2.1

MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 29.1 on the Noometry Index.

Last verified . 17 shared benchmarks.

GPT-4 OpenAI

29.1

Rank #316 Confirmed

MiniMax-M2.1 MiniMax

38.9

Rank #178 Confirmed

Summary

  • They share 17 benchmarks with published results for both. GPT-4 scores higher in 1 category and MiniMax-M2.1 in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where MiniMax-M2.1 leads 38.3 to 10.8.
  • MiniMax-M2.1 is cheaper at $0.30 / $1.20 per million input/output tokens, against $30 / $60 for GPT-4.
  • MiniMax-M2.1 accepts more context: 205K tokens versus 8K.
  • MiniMax-M2.1 has downloadable open weights; the other is API-only.

Side by side

GPT-4MiniMax-M2.1
ProviderOpenAIMiniMax
Noometry Index29.138.9
Released2023-03-142025-12-23
WeightsProprietaryOpen
Context window8K205K
Max output8K131K
Input $ / M tokens$30$0.30
Output $ / M tokens$60$1.20
Results tracked3822

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

Category by category

Coding MiniMax-M2.1 leads

GPT-4: 31.6 (#283), MiniMax-M2.1: 40.4 (#143)

BenchmarkGPT-4MiniMax-M2.1
LMArena Coding12541421
LMArena WebDev—1384
WeirdML12.4%—
BigCodeBench Instruct46%—
BigCodeBench Complete57.2%—
ALE-Bench—623.83
HumanEval+79.3%—

Agentic & Tool Use Not comparable

GPT-4: —, MiniMax-M2.1: 27.9 (#98)

BenchmarkGPT-4MiniMax-M2.1
Terminal-Bench—36.6%
METR Time Horizons36.1%—

Reasoning GPT-4 leads

GPT-4: 17.8 (#289), MiniMax-M2.1: 16.6 (#302)

BenchmarkGPT-4MiniMax-M2.1
LMArena Hard Prompts12411411
NYT Connections (extended)—11.2%
Chess Puzzles4%—
Mystery Game Puzzles12%—
DTBench62.7%—
LMCA17.1%—
BIG-Bench Hard75.1%—
Epoch Capabilities Index125.89—
ForecastBench57.8—
HellaSwag95.3%—
WinoGrande87.5%—

Math MiniMax-M2.1 leads

GPT-4: 10.8 (#309), MiniMax-M2.1: 38.3 (#138)

BenchmarkGPT-4MiniMax-M2.1
LMArena Math12691397
OTIS Mock AIME 2024-20251.1%—
MATH Level 523%—
GSM8K92%—

Knowledge MiniMax-M2.1 leads

GPT-4: 18.4 (#282), MiniMax-M2.1: 38.3 (#147)

BenchmarkGPT-4MiniMax-M2.1
LMArena Expert12111431
GPQA Diamond35.7%—
Vectara Hallucination Rate—11.8%
MMLU86.4%—
TriviaQA84.8%—

Multilingual MiniMax-M2.1 leads

GPT-4: 40.6 (#215), MiniMax-M2.1: 50.0 (#128)

BenchmarkGPT-4MiniMax-M2.1
LMArena Non-English12461378
LMArena Chinese12421430
LMArena French12831404
LMArena German12511381
LMArena Japanese12091287
LMArena Korean11841298
LMArena Russian12511387
LMArena Spanish12611397

Instruction Following MiniMax-M2.1 leads

GPT-4: 65.3 (#222), MiniMax-M2.1: 73.8 (#112)

BenchmarkGPT-4MiniMax-M2.1
LMArena Instruction Following12411400

Long Context MiniMax-M2.1 leads

GPT-4: 37.7 (#212), MiniMax-M2.1: 43.2 (#101)

BenchmarkGPT-4MiniMax-M2.1
LMArena Longer Query12441416

Writing & Preference MiniMax-M2.1 leads

GPT-4: 34.9 (#268), MiniMax-M2.1: 58.3 (#120)

BenchmarkGPT-4MiniMax-M2.1
LMArena Text12631392
LMArena Creative Writing12441361
LMArena Multi-Turn12571396
EQ-Bench Creative Writing752—

Frequently asked questions

Is GPT-4 better than MiniMax-M2.1?

MiniMax-M2.1 is the stronger model overall, scoring 38.9 to 29.1 on the Noometry Index.

Which is cheaper, GPT-4 or MiniMax-M2.1?

MiniMax-M2.1 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-4 lists at $30 and $60.

Is GPT-4 or MiniMax-M2.1 better for coding?

MiniMax-M2.1 scores higher on coding benchmarks: 40.4 versus 31.6 in the Noometry coding category.

Which has the bigger context window?

MiniMax-M2.1 does, with 205K tokens against 8K.

How many benchmarks do GPT-4 and MiniMax-M2.1 share?

17 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and MiniMax-M2.1 has 22.

Related comparisons

Go deeper