Noometry

Model comparison

GPT-4o vs MiniMax-M2.7

MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 28.6 on the Noometry Index.

Last verified . 21 shared benchmarks.

GPT-4o OpenAI

28.6

Rank #324 Confirmed

MiniMax-M2.7 MiniMax

37.7

Rank #196 Confirmed

Summary

  • They share 21 benchmarks with published results for both. GPT-4o scores higher in 0 categories and MiniMax-M2.7 in 9 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in coding, where MiniMax-M2.7 leads 41.8 to 24.8.
  • The biggest single-benchmark swing is WeirdML: 25.1% for GPT-4o and 37% for MiniMax-M2.7.
  • MiniMax-M2.7 is cheaper at $0.30 / $1.20 per million input/output tokens, against $2.50 / $10 for GPT-4o.
  • MiniMax-M2.7 accepts more context: 205K tokens versus 128K.
  • MiniMax-M2.7 has downloadable open weights; the other is API-only.

Side by side

GPT-4oMiniMax-M2.7
ProviderOpenAIMiniMax
Noometry Index28.637.7
Released2024-05-132026-03-18
WeightsProprietaryOpen
Context window128K205K
Max output16K131K
Input $ / M tokens$2.50$0.30
Output $ / M tokens$10$1.20
Results tracked7230

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

Category by category

Coding MiniMax-M2.7 leads

GPT-4o: 24.8 (#328), MiniMax-M2.7: 41.8 (#120)

BenchmarkGPT-4oMiniMax-M2.7
WeirdML25.1%37%
LMArena Coding12971454
SWE-bench Verified31%—
SWE-bench Verified (bash only)21.6%—
Aider Polyglot45.3%—
LMArena WebDev—1398
SciCode—47%
GSO0%—
BigCodeBench Instruct51.1%—
LiveBench Coding51.4%—
BigCodeBench Complete61.1%—
CadEval26%—
ALE-Bench—599.25
HumanEval+87.2%—
MBPP+72.2%—

Agentic & Tool Use MiniMax-M2.7 leads

GPT-4o: 21.0 (#141), MiniMax-M2.7: 25.1 (#111)

BenchmarkGPT-4oMiniMax-M2.7
Terminal-Bench—45.1%
GDPval9.9%—
TheAgentCompany8.6%—
Cybench12.5%—
BALROG32.3%—
ExploitBench—13.3%
GBAEval—0%
LMArena Search1006—
METR Time Horizons40.8%—

Reasoning MiniMax-M2.7 leads

GPT-4o: 9.4 (#343), MiniMax-M2.7: 19.7 (#253)

BenchmarkGPT-4oMiniMax-M2.7
CritPt0%0.6%
LMArena Hard Prompts12811422
Epoch Capabilities Index128.97145.85
ARC-AGI-20%—
SimpleBench17.8%—
NYT Connections (extended)—24.7%
ARC-AGI-14.5%—
Chess Puzzles13%—
EnigmaEval0.8%—
Thematic Generalization—39.3%
LiveBench Reasoning55.8%—
DTBench64.5%—
LiveBench Data Analysis60.9%—
LMCA16.6%—
ForecastBench57.7—
LiveBench55.3%—

Math MiniMax-M2.7 leads

GPT-4o: 10.6 (#312), MiniMax-M2.7: 25.9 (#263)

BenchmarkGPT-4oMiniMax-M2.7
LMArena Math12851420
FrontierMath (Tiers 1-3)0.4%—
OTIS Mock AIME 2024-20256.4%—
ProofBench—3%
Omni-MATH29.3%—
LiveBench Math49.5%—
MATH Level 553.3%—
FrontierMath (Feb 2025 set)0.3%—

Knowledge MiniMax-M2.7 leads

GPT-4o: 28.8 (#242), MiniMax-M2.7: 37.7 (#152)

BenchmarkGPT-4oMiniMax-M2.7
Vectara Hallucination Rate9.6%12.9%
LMArena Expert12501444
GPQA Diamond49.2%—
Humanity's Last Exam2.7%—
SimpleQA Verified26%—
MMLU-Pro71.3%—
Confabulations15.3%—
GPQA (HELM)52%—
MMLU88.1%—

Multimodal Not comparable

GPT-4o: 34.5 (#91), MiniMax-M2.7: —

BenchmarkGPT-4oMiniMax-M2.7
LMArena Vision1137—
Video-MME71.9%—
GeoBench71%—
VPCT40%—
ScienceQA88.5%—

Multilingual MiniMax-M2.7 leads

GPT-4o: 43.2 (#186), MiniMax-M2.7: 50.3 (#123)

BenchmarkGPT-4oMiniMax-M2.7
LMArena Non-English12831382
LMArena Chinese12771441
LMArena French13041421
LMArena German12821398
LMArena Japanese12571262
LMArena Korean12341313
LMArena Russian12861383
LMArena Spanish12921403

Instruction Following MiniMax-M2.7 leads

GPT-4o: 66.6 (#207), MiniMax-M2.7: 74.1 (#103)

BenchmarkGPT-4oMiniMax-M2.7
LMArena Instruction Following12781405
LiveBench Instruction Following68.6%—
IFEval81.7%—

Long Context MiniMax-M2.7 leads

GPT-4o: 39.4 (#179), MiniMax-M2.7: 43.3 (#99)

BenchmarkGPT-4oMiniMax-M2.7
LMArena Longer Query12891419
Fiction.LiveBench66.7%—

Writing & Preference MiniMax-M2.7 leads

GPT-4o: 52.6 (#166), MiniMax-M2.7: 58.9 (#112)

BenchmarkGPT-4oMiniMax-M2.7
LMArena Text13001405
LMArena Creative Writing12921354
LMArena Multi-Turn13021412
Short-Story Creative Writing81.8%—
WildBench82.8%—
LiveBench Language47.6%—

Frequently asked questions

Is GPT-4o better than MiniMax-M2.7?

MiniMax-M2.7 is the stronger model overall, scoring 37.7 to 28.6 on the Noometry Index.

Which is cheaper, GPT-4o or MiniMax-M2.7?

MiniMax-M2.7 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-4o lists at $2.50 and $10.

Is GPT-4o or MiniMax-M2.7 better for coding?

MiniMax-M2.7 scores higher on coding benchmarks: 41.8 versus 24.8 in the Noometry coding category.

Which has the bigger context window?

MiniMax-M2.7 does, with 205K tokens against 128K.

How many benchmarks do GPT-4o and MiniMax-M2.7 share?

21 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and MiniMax-M2.7 has 30.

Related comparisons

Go deeper