Noometry

Model comparison

GPT-5.5 vs MiniMax-M3

GPT-5.5 is the stronger model overall, scoring 63.4 to 43.8 on the Noometry Index. MiniMax-M3 costs 21× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Last verified . 41 shared benchmarks.

GPT-5.5 OpenAI

63.4

Rank #9 Confirmed

MiniMax-M3 MiniMax

43.8

Rank #85 Confirmed

Summary

  • They share 41 benchmarks with published results for both. GPT-5.5 scores higher in 10 categories and MiniMax-M3 in 0 categories; 10 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where GPT-5.5 leads 72.8 to 30.1.
  • The biggest single-benchmark swing is GBAEval: 53.2% for GPT-5.5 and 0.9% for MiniMax-M3.
  • MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $5 / $30 for GPT-5.5.
  • GPT-5.5 accepts more context: 1.05M tokens versus 1M.
  • MiniMax-M3 has downloadable open weights; the other is API-only.

Side by side

GPT-5.5MiniMax-M3
ProviderOpenAIMiniMax
Noometry Index63.443.8
Released2026-04-232026-06-01
WeightsProprietaryOpen
Context window1.05M1M
Max output128K512K
Input $ / M tokens$5$0.30
Output $ / M tokens$30$1.20
Results tracked7141

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

Category by category

Coding GPT-5.5 leads

GPT-5.5: 58.2 (#17), MiniMax-M3: 41.8 (#118)

BenchmarkGPT-5.5MiniMax-M3
FrontierCode43%14.7%
LMArena WebDev15131482
SciCode56.1%47.1%
LMArena Coding14941469
ALE-Bench1,943640.02
SWE-bench Verified80.6%—
DeepSWE67%—
GSO40.2%—
WeirdML84.9%—
MirrorCode10%—

Agentic & Tool Use GPT-5.5 leads

GPT-5.5: 50.7 (#6), MiniMax-M3: 22.6 (#130)

BenchmarkGPT-5.5MiniMax-M3
APEX-Agents55.1%37.7%
OSWorld 2.013%4.6%
GBAEval53.2%0.9%
Vending-Bench 27,5242,158
Terminal-Bench84.7%—
Remote Labor Index6.3%—
τ²-bench Banking44.6%—
DeepResearch Bench54%—
PostTrainBench27.2%—
ExploitBench47.4%—
GDP.pdf26%—
LMArena Search1242—

Reasoning GPT-5.5 leads

GPT-5.5: 72.8 (#11), MiniMax-M3: 30.1 (#87)

BenchmarkGPT-5.5MiniMax-M3
SimpleBench69%45.8%
NYT Connections (extended)96.2%65.1%
CritPt27.1%3.7%
Chess Puzzles54%14%
LMArena Hard Prompts14891447
Mystery Game Puzzles56%8%
DTBench96%78.9%
LMCA54.3%33.7%
Surface Evolver Bench88.1%55%
Epoch Capabilities Index159.1146.95
ForecastBench60.661.4
ARC-AGI-285%—
Kagi LLM Benchmark88.8%—
ARC-AGI-195%—
EBR-Bench34.3%—
Bench to the Future 30.14—

Math GPT-5.5 leads

GPT-5.5: 81.7 (#11), MiniMax-M3: 40.0 (#95)

BenchmarkGPT-5.5MiniMax-M3
OTIS Mock AIME 2024-2025100%71.1%
ProofBench50%18%
LMArena Math14861429
FrontierMath (Tiers 1-3)85.3%—
FrontierMath Tier 472.5%—
MathArena Final-Answer Competitions94.3%—
FrontierMath (Feb 2025 set)51.7%—
FrontierMath Erdős0%—
FrontierMath Tier 4 (v1)35.4%—

Knowledge GPT-5.5 leads

GPT-5.5: 64.4 (#17), MiniMax-M3: 58.4 (#35)

BenchmarkGPT-5.5MiniMax-M3
GPQA Diamond94%90.9%
LMArena Expert15081461
SimpleQA Verified63%—
Vectara Hallucination Rate9.3%—

Multimodal GPT-5.5 leads

GPT-5.5: 46.9 (#12), MiniMax-M3: 40.2 (#51)

BenchmarkGPT-5.5MiniMax-M3
LMArena Vision12971253
LMArena Document14861435
Blueprint-Bench 236.2%—
Furniture Assembly44.2%—

Multilingual GPT-5.5 leads

GPT-5.5: 56.4 (#20), MiniMax-M3: 53.0 (#75)

BenchmarkGPT-5.5MiniMax-M3
LMArena Non-English14671420
LMArena Chinese15331463
LMArena French14861447
LMArena German14801426
LMArena Japanese14981381
LMArena Korean14601372
LMArena Russian14731428
LMArena Spanish14681432

Instruction Following GPT-5.5 leads

GPT-5.5: 77.5 (#18), MiniMax-M3: 75.5 (#62)

BenchmarkGPT-5.5MiniMax-M3
LMArena Instruction Following14791433

Long Context GPT-5.5 leads

GPT-5.5: 48.3 (#12), MiniMax-M3: 44.2 (#72)

BenchmarkGPT-5.5MiniMax-M3
LMArena Longer Query14841445
CL-bench Life22.2%—

Writing & Preference GPT-5.5 leads

GPT-5.5: 72.7 (#13), MiniMax-M3: 62.1 (#83)

BenchmarkGPT-5.5MiniMax-M3
LMArena Text14721433
LMArena Creative Writing14551404
EQ-Bench 413151150
LMArena Multi-Turn14761442
EQ-Bench Creative Writing1844—

Frequently asked questions

Is GPT-5.5 better than MiniMax-M3?

GPT-5.5 is the stronger model overall, scoring 63.4 to 43.8 on the Noometry Index. MiniMax-M3 costs 21× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

Which is cheaper, GPT-5.5 or MiniMax-M3?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-5.5 lists at $5 and $30.

Is GPT-5.5 or MiniMax-M3 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.2 versus 41.8 in the Noometry coding category.

Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 1M.

How many benchmarks do GPT-5.5 and MiniMax-M3 share?

41 benchmarks have published results for both models. GPT-5.5 has 71 scored results on Noometry and MiniMax-M3 has 41.

Related comparisons

Go deeper