Model comparison

GPT-4 Turbo vs MiniMax-M3

MiniMax-M3 is the stronger model overall, scoring 43.5 to 30.3 on the Noometry Index.

Last verified October 10, 2026. 26 shared benchmarks.

GPT-4 Turbo OpenAI

30.3

Rank #281 Confirmed

MiniMax-M3

MiniMax

43.5

Rank #81 Confirmed

Summary

They share 26 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and MiniMax-M3 in 9 categories; 9 gaps are clear of the uncertainty.
The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 24.3.
The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 71.1% for MiniMax-M3.
MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
MiniMax-M3 accepts more context: 1M tokens versus 128K.
MiniMax-M3 has downloadable open weights; the other is API-only.

Side by side

GPT-4 Turbo and MiniMax-M3 specifications
	GPT-4 Turbo	MiniMax-M3
Provider	OpenAI	MiniMax
Noometry Index	30.3	43.5
Released	2023-11-06	2026-06-01
Weights	Proprietary	Open
Context window	128K	1M
Max output	4K	512K
Input $ / M tokens	$10	$0.30
Output $ / M tokens	$30	$1.20
Results tracked	36	41

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

Category by category

Coding MiniMax-M3 leads

GPT-4 Turbo: 33.4 (#247), MiniMax-M3: 41.5 (#111)

Coding benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Coding	1268	1470
FrontierCode	—	14.7%
LMArena WebDev	—	1482
SciCode	—	47.1%
WeirdML	18%	—
BigCodeBench Instruct	48.2%	—
BigCodeBench Complete	58.2%	—
ALE-Bench	—	640.02
HumanEval+	86.6%	—
MBPP+	73.3%	—

Agentic & Tool Use Not comparable

GPT-4 Turbo: —, MiniMax-M3: 22.5 (#121)

Agentic & Tool Use benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
APEX-Agents	—	37.7%
OSWorld 2.0	—	4.6%
GBAEval	—	0.9%
METR Time Horizons	36.7%	—
Vending-Bench 2	—	2,158

Reasoning MiniMax-M3 leads

GPT-4 Turbo: 14.9 (#311), MiniMax-M3: 29.3 (#83)

Reasoning benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
SimpleBench	25.1%	45.8%
Chess Puzzles	6%	14%
LMArena Hard Prompts	1251	1447
DTBench	61.6%	78.9%
LMCA	9.8%	33.7%
Epoch Capabilities Index	127.25	147
ForecastBench	59.4	61.4
NYT Connections (extended)	—	65.1%
CritPt	—	3.7%
Mystery Game Puzzles	—	8%
Surface Evolver Bench	—	55%

Math MiniMax-M3 leads

GPT-4 Turbo: 8.6 (#316), MiniMax-M3: 39.1 (#83)

Math benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
OTIS Mock AIME 2024-2025	6.7%	71.1%
LMArena Math	1272	1432
FrontierMath (Tiers 1-3)	0.7%	—
ProofBench	—	18%
MATH Level 5	46.7%	—

Knowledge MiniMax-M3 leads

GPT-4 Turbo: 24.3 (#263), MiniMax-M3: 58.4 (#32)

Knowledge benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
GPQA Diamond	46.6%	90.9%
LMArena Expert	1223	1462
Confabulations	28.4%	—
MMLU	81.3%	—

Multimodal MiniMax-M3 leads

GPT-4 Turbo: 30.0 (#104), MiniMax-M3: 39.5 (#45)

Multimodal benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Vision	1090	1254
LMArena Document	—	1435

Multilingual MiniMax-M3 leads

GPT-4 Turbo: 40.7 (#211), MiniMax-M3: 53.2 (#70)

Multilingual benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Non-English	1245	1420
LMArena Chinese	1242	1465
LMArena French	1275	1445
LMArena German	1259	1424
LMArena Japanese	1194	1381
LMArena Korean	1187	1372
LMArena Russian	1259	1430
LMArena Spanish	1260	1431

Instruction Following MiniMax-M3 leads

GPT-4 Turbo: 66.0 (#211), MiniMax-M3: 75.7 (#56)

Instruction Following benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Instruction Following	1250	1433

Long Context MiniMax-M3 leads

GPT-4 Turbo: 38.0 (#200), MiniMax-M3: 44.2 (#68)

Long Context benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Longer Query	1254	1445

Writing & Preference MiniMax-M3 leads

GPT-4 Turbo: 47.9 (#201), MiniMax-M3: 62.3 (#79)

Writing & Preference benchmarks
Benchmark	GPT-4 Turbo	MiniMax-M3
LMArena Text	1272	1433
LMArena Creative Writing	1269	1405
LMArena Multi-Turn	1267	1441
EQ-Bench 4	—	1150

Frequently asked questions

Is GPT-4 Turbo better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 30.3 on the Noometry Index.

Which is cheaper, GPT-4 Turbo or MiniMax-M3?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; GPT-4 Turbo lists at $10 and $30.

Is GPT-4 Turbo or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 33.4 in the Noometry coding category.

Which has the bigger context window?

MiniMax-M3 does, with 1M tokens against 128K.

How many benchmarks do GPT-4 Turbo and MiniMax-M3 share?

26 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and MiniMax-M3 has 41.