Model comparison

DeepSeek-V3.2-Speciale vs MiniMax-M3

MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.7 on the Noometry Index.

Last verified October 12, 2026. 1 shared benchmarks.

DeepSeek-V3.2-Speciale

DeepSeek

39.7

Rank #162 Reported

MiniMax-M3

MiniMax

43.8

Rank #85 Confirmed

Summary

They share 1 benchmark with published results for both. DeepSeek-V3.2-Speciale scores higher in 1 category and MiniMax-M3 in 2 categories; 3 gaps are clear of the uncertainty.
The widest gap is in writing & preference, where MiniMax-M3 leads 62.1 to 46.0.
The biggest single-benchmark swing is SimpleBench: 52.6% for DeepSeek-V3.2-Speciale and 45.8% for MiniMax-M3.
MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.58 / $1.68 for DeepSeek-V3.2-Speciale.
MiniMax-M3 accepts more context: 1M tokens versus 128K.

Side by side

DeepSeek-V3.2-Speciale and MiniMax-M3 specifications
	DeepSeek-V3.2-Speciale	MiniMax-M3
Provider	DeepSeek	MiniMax
Noometry Index	39.7	43.8
Released	2025-12-01	2026-06-01
Weights	Open	Open
Context window	128K	1M
Max output	128K	512K
Input $ / M tokens	$0.58	$0.30
Output $ / M tokens	$1.68	$1.20
Results tracked	3	41

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

Category by category

Coding MiniMax-M3 leads

DeepSeek-V3.2-Speciale: 40.4 (#140), MiniMax-M3: 41.8 (#118)

Coding benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
FrontierCode	—	14.7%
LMArena WebDev	—	1482
SciCode	—	47.1%
WeirdML	46.7%	—
LMArena Coding	—	1469
ALE-Bench	—	640.02

Agentic & Tool Use Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 22.6 (#130)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
APEX-Agents	—	37.7%
OSWorld 2.0	—	4.6%
GBAEval	—	0.9%
Vending-Bench 2	—	2,158

Reasoning DeepSeek-V3.2-Speciale leads

DeepSeek-V3.2-Speciale: 32.9 (#73), MiniMax-M3: 30.1 (#87)

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
SimpleBench	52.6%	45.8%
NYT Connections (extended)	—	65.1%
CritPt	—	3.7%
Chess Puzzles	—	14%
LMArena Hard Prompts	—	1447
Mystery Game Puzzles	—	8%
DTBench	—	78.9%
LMCA	—	33.7%
Surface Evolver Bench	—	55%
Epoch Capabilities Index	—	146.95
ForecastBench	—	61.4

Math Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 40.0 (#95)

Math benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
OTIS Mock AIME 2024-2025	—	71.1%
ProofBench	—	18%
LMArena Math	—	1429

Knowledge Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 58.4 (#35)

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
GPQA Diamond	—	90.9%
LMArena Expert	—	1461

Multimodal Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 40.2 (#51)

Multimodal benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
LMArena Vision	—	1253
LMArena Document	—	1435

Multilingual Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 53.0 (#75)

Multilingual benchmarks
Benchmark	DeepSeek-V3.2-Speciale	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

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 75.5 (#62)

Instruction Following benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
LMArena Instruction Following	—	1433

Long Context Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M3: 44.2 (#72)

Long Context benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
LMArena Longer Query	—	1445

Writing & Preference MiniMax-M3 leads

DeepSeek-V3.2-Speciale: 46.0 (#222), MiniMax-M3: 62.1 (#83)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M3
LMArena Text	—	1433
LMArena Creative Writing	—	1404
EQ-Bench Creative Writing	1276	—
EQ-Bench 4	—	1150
LMArena Multi-Turn	—	1442

Frequently asked questions

Is DeepSeek-V3.2-Speciale better than MiniMax-M3?

MiniMax-M3 is the stronger model overall, scoring 43.8 to 39.7 on the Noometry Index.

Which is cheaper, DeepSeek-V3.2-Speciale or MiniMax-M3?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; DeepSeek-V3.2-Speciale lists at $0.58 and $1.68.

Is DeepSeek-V3.2-Speciale or MiniMax-M3 better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.8 versus 40.4 in the Noometry coding category.

Which has the bigger context window?

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

How many benchmarks do DeepSeek-V3.2-Speciale and MiniMax-M3 share?

1 benchmark has published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and MiniMax-M3 has 41.