Model comparison

Llama 3.2 90B vs MiniMax-M3

MiniMax-M3 is the stronger model overall, scoring 43.8 to 27.5 on the Noometry Index.

Last verified October 10, 2026. 4 shared benchmarks.

Summary

They share 4 benchmarks with published results for both. Llama 3.2 90B scores higher in 1 category and MiniMax-M3 in 4 categories; 5 gaps are clear of the uncertainty.
The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 21.7.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 2.6% for Llama 3.2 90B and 71.1% for MiniMax-M3.

Side by side

Llama 3.2 90B and MiniMax-M3 specifications
	Llama 3.2 90B	MiniMax-M3
Provider	Meta	MiniMax
Noometry Index	27.5	43.8
Released	2024-09-24	2026-06-01
Weights	Open	Open
Context window	—	1M
Max output	—	512K
Input $ / M tokens	—	$0.30
Output $ / M tokens	—	$1.20
Results tracked	9	41

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

Category by category

Coding Not comparable

Llama 3.2 90B: —, MiniMax-M3: 41.8 (#118)

Coding benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
FrontierCode	—	14.7%
LMArena WebDev	—	1482
SciCode	—	47.1%
LMArena Coding	—	1469
ALE-Bench	—	640.02

Agentic & Tool Use Llama 3.2 90B leads

Llama 3.2 90B: 30.0 (#80), MiniMax-M3: 22.6 (#130)

Agentic & Tool Use benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
APEX-Agents	—	37.7%
OSWorld 2.0	—	4.6%
BALROG	27.3%	—
GBAEval	—	0.9%
Vending-Bench 2	—	2,158

Reasoning MiniMax-M3 leads

Llama 3.2 90B: 21.7 (#217), MiniMax-M3: 30.1 (#87)

Reasoning benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
Epoch Capabilities Index	125.5	146.95
SimpleBench	—	45.8%
NYT Connections (extended)	—	65.1%
CritPt	—	3.7%
Chess Puzzles	—	14%
EnigmaEval	0.4%	—
LMArena Hard Prompts	—	1447
Mystery Game Puzzles	—	8%
DTBench	—	78.9%
LMCA	—	33.7%
Surface Evolver Bench	—	55%
ForecastBench	—	61.4

Math MiniMax-M3 leads

Llama 3.2 90B: 11.1 (#308), MiniMax-M3: 40.0 (#95)

Math benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
OTIS Mock AIME 2024-2025	2.6%	71.1%
ProofBench	—	18%
LMArena Math	—	1429
MATH Level 5	39.4%	—

Knowledge MiniMax-M3 leads

Llama 3.2 90B: 21.7 (#274), MiniMax-M3: 58.4 (#35)

Knowledge benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
GPQA Diamond	41%	90.9%
LMArena Expert	—	1461
MMLU	80.3%	—

Multimodal MiniMax-M3 leads

Llama 3.2 90B: 25.4 (#124), MiniMax-M3: 40.2 (#51)

Multimodal benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
LMArena Vision	1000	1253
GeoBench	52%	—
LMArena Document	—	1435

Multilingual Not comparable

Llama 3.2 90B: —, MiniMax-M3: 53.0 (#75)

Multilingual benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
LMArena Non-English	—	1420
LMArena Chinese	—	1463
LMArena French	—	1447
LMArena German	—	1426
LMArena Japanese	—	1381
LMArena Korean	—	1372
LMArena Russian	—	1428
LMArena Spanish	—	1432

Instruction Following Not comparable

Llama 3.2 90B: —, MiniMax-M3: 75.5 (#62)

Instruction Following benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
LMArena Instruction Following	—	1433

Long Context Not comparable

Llama 3.2 90B: —, MiniMax-M3: 44.2 (#72)

Long Context benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
LMArena Longer Query	—	1445

Writing & Preference Not comparable

Llama 3.2 90B: —, MiniMax-M3: 62.1 (#83)

Writing & Preference benchmarks
Benchmark	Llama 3.2 90B	MiniMax-M3
LMArena Text	—	1433
LMArena Creative Writing	—	1404
EQ-Bench 4	—	1150
LMArena Multi-Turn	—	1442

Frequently asked questions

Is Llama 3.2 90B better than MiniMax-M3?