Model comparison

MiniMax-M3 vs Qwen3.5 27B

MiniMax-M3 is the stronger model overall, scoring 43.5 to 41.4 on the Noometry Index.

Last verified October 10, 2026. 24 shared benchmarks.

MiniMax-M3

MiniMax

43.5

Rank #81 Confirmed

Qwen3.5 27B

Alibaba (Qwen)

41.4

Rank #127 Confirmed

Summary

They share 24 benchmarks with published results for both. MiniMax-M3 scores higher in 9 categories and Qwen3.5 27B in 0 categories; 8 gaps are clear of the uncertainty.
The widest gap is in knowledge, where MiniMax-M3 leads 58.4 to 37.4.
The biggest single-benchmark swing is NYT Connections (extended): 65.1% for MiniMax-M3 and 47.9% for Qwen3.5 27B.
MiniMax-M3 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
MiniMax-M3 accepts more context: 1M tokens versus 262K.

Side by side

MiniMax-M3 and Qwen3.5 27B specifications
	MiniMax-M3	Qwen3.5 27B
Provider	MiniMax	Alibaba (Qwen)
Noometry Index	43.5	41.4
Released	2026-06-01	2026-02-23
Weights	Open	Open
Context window	1M	262K
Max output	512K	66K
Input $ / M tokens	$0.30	$0.30
Output $ / M tokens	$1.20	$2.40
Results tracked	41	28

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

Category by category

Coding MiniMax-M3 leads

MiniMax-M3: 41.5 (#111), Qwen3.5 27B: 38.6 (#159)

Coding benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena WebDev	1482	1357
LMArena Coding	1470	1427
ALE-Bench	640.02	349.45
FrontierCode	14.7%	—
SciCode	47.1%	—
WeirdML	—	39.5%

Agentic & Tool Use Not comparable

MiniMax-M3: 22.5 (#121), Qwen3.5 27B: —

Agentic & Tool Use benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
Vending-Bench 2	2,158	201.98
APEX-Agents	37.7%	—
OSWorld 2.0	4.6%	—
GBAEval	0.9%	—

Reasoning MiniMax-M3 leads

MiniMax-M3: 29.3 (#83), Qwen3.5 27B: 26.6 (#116)

Reasoning benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
NYT Connections (extended)	65.1%	47.9%
LMArena Hard Prompts	1447	1414
DTBench	78.9%	82.4%
LMCA	33.7%	34%
SimpleBench	45.8%	—
CritPt	3.7%	—
Chess Puzzles	14%	—
Thematic Generalization	—	45.5%
Mystery Game Puzzles	8%	—
Surface Evolver Bench	55%	—
Epoch Capabilities Index	147	—
ForecastBench	61.4	—

Math MiniMax-M3 leads

MiniMax-M3: 39.1 (#83), Qwen3.5 27B: 36.8 (#103)

Math benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Math	1432	1428
MathArena Final-Answer Competitions	—	56.7%
OTIS Mock AIME 2024-2025	71.1%	—
ProofBench	18%	—

Knowledge MiniMax-M3 leads

MiniMax-M3: 58.4 (#32), Qwen3.5 27B: 37.4 (#149)

Knowledge benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Expert	1462	1428
GPQA Diamond	90.9%	—
Vectara Hallucination Rate	—	12.1%

Multimodal Too close to call

MiniMax-M3: 39.5 (#45), Qwen3.5 27B: 38.7 (#56)

Multimodal benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Vision	1254	1240
LMArena Document	1435	—

Multilingual MiniMax-M3 leads

MiniMax-M3: 53.2 (#70), Qwen3.5 27B: 51.0 (#112)

Multilingual benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Non-English	1420	1390
LMArena Chinese	1465	1478
LMArena French	1445	1409
LMArena German	1424	1393
LMArena Japanese	1381	1345
LMArena Korean	1372	1359
LMArena Russian	1430	1389
LMArena Spanish	1431	1408

Instruction Following MiniMax-M3 leads

MiniMax-M3: 75.7 (#56), Qwen3.5 27B: 73.7 (#114)

Instruction Following benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Instruction Following	1433	1393

Long Context MiniMax-M3 leads

MiniMax-M3: 44.2 (#68), Qwen3.5 27B: 43.1 (#103)

Long Context benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Longer Query	1445	1413

Writing & Preference MiniMax-M3 leads

MiniMax-M3: 62.3 (#79), Qwen3.5 27B: 59.5 (#106)

Writing & Preference benchmarks
Benchmark	MiniMax-M3	Qwen3.5 27B
LMArena Text	1433	1409
LMArena Creative Writing	1405	1362
LMArena Multi-Turn	1441	1410
EQ-Bench 4	1150	—

Frequently asked questions

Is MiniMax-M3 better than Qwen3.5 27B?

MiniMax-M3 is the stronger model overall, scoring 43.5 to 41.4 on the Noometry Index.

Which is cheaper, MiniMax-M3 or Qwen3.5 27B?

MiniMax-M3 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen3.5 27B lists at $0.30 and $2.40.

Is MiniMax-M3 or Qwen3.5 27B better for coding?

MiniMax-M3 scores higher on coding benchmarks: 41.5 versus 38.6 in the Noometry coding category.

Which has the bigger context window?

MiniMax-M3 does, with 1M tokens against 262K.

How many benchmarks do MiniMax-M3 and Qwen3.5 27B share?

24 benchmarks have published results for both models. MiniMax-M3 has 41 scored results on Noometry and Qwen3.5 27B has 28.