Model comparison

DeepSeek-V3.2-Speciale vs MiniMax-M2

DeepSeek-V3.2-Speciale is the stronger model overall, scoring 39.1 to 36.9 on the Noometry Index. MiniMax-M2 costs 1.6× less per token, which makes it the better buy when DeepSeek-V3.2-Speciale's lead doesn't matter for your workload.

Last verified October 10, 2026. 0 shared benchmarks.

DeepSeek-V3.2-Speciale

DeepSeek

39.1

Rank #159 Reported

MiniMax-M2

MiniMax

36.9

Rank #201 Confirmed

Summary

The widest gap is in reasoning, where DeepSeek-V3.2-Speciale leads 32.1 to 18.6.
MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.58 / $1.68 for DeepSeek-V3.2-Speciale.
MiniMax-M2 accepts more context: 205K tokens versus 128K.

Side by side

DeepSeek-V3.2-Speciale and MiniMax-M2 specifications
	DeepSeek-V3.2-Speciale	MiniMax-M2
Provider	DeepSeek	MiniMax
Noometry Index	39.1	36.9
Released	2025-12-01	2025-10-27
Weights	Open	Open
Context window	128K	205K
Max output	128K	131K
Input $ / M tokens	$0.58	$0.30
Output $ / M tokens	$1.68	$1.20
Results tracked	3	21

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

Category by category

Coding Too close to call

DeepSeek-V3.2-Speciale: 39.9 (#136), MiniMax-M2: 39.1 (#151)

Coding benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
SWE-bench Verified (bash only)	—	61%
LMArena WebDev	—	1297
WeirdML	46.7%	—
LMArena Coding	—	1370

Agentic & Tool Use Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 24.9 (#102)

Agentic & Tool Use benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
Terminal-Bench	—	30%
Vending-Bench 2	—	160.6

Reasoning DeepSeek-V3.2-Speciale leads

DeepSeek-V3.2-Speciale: 32.1 (#65), MiniMax-M2: 18.6 (#256)

Reasoning benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
SimpleBench	52.6%	—
Kagi LLM Benchmark	—	57.8%
NYT Connections (extended)	—	14.8%
LMArena Hard Prompts	—	1357

Math Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 34.6 (#158)

Math benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Math	—	1350

Knowledge Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 36.3 (#160)

Knowledge benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Expert	—	1336

Multilingual Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 45.6 (#165)

Multilingual benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Non-English	—	1314
LMArena Chinese	—	1367
LMArena French	—	1336
LMArena German	—	1366
LMArena Russian	—	1332
LMArena Spanish	—	1326

Instruction Following Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 70.4 (#160)

Instruction Following benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Instruction Following	—	1329

Long Context Not comparable

DeepSeek-V3.2-Speciale: —, MiniMax-M2: 40.5 (#148)

Long Context benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Longer Query	—	1332

Writing & Preference MiniMax-M2 leads

DeepSeek-V3.2-Speciale: 46.1 (#217), MiniMax-M2: 53.3 (#157)

Writing & Preference benchmarks
Benchmark	DeepSeek-V3.2-Speciale	MiniMax-M2
LMArena Text	—	1340
LMArena Creative Writing	—	1288
EQ-Bench Creative Writing	1276	—
LMArena Multi-Turn	—	1362

Frequently asked questions

Is DeepSeek-V3.2-Speciale better than MiniMax-M2?

Which is cheaper, DeepSeek-V3.2-Speciale or MiniMax-M2?

MiniMax-M2 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-M2 better for coding?

They score almost the same on coding (39.9 vs 39.1); test both on your own repository before choosing.

Which has the bigger context window?

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

How many benchmarks do DeepSeek-V3.2-Speciale and MiniMax-M2 share?

0 benchmarks have published results for both models. DeepSeek-V3.2-Speciale has 3 scored results on Noometry and MiniMax-M2 has 21.