Model comparison

MiniMax-M2.5 vs Qwen3.5 27B

Qwen3.5 27B is the stronger model overall, scoring 41.4 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 1.6× less per token, which makes it the better buy when Qwen3.5 27B's lead doesn't matter for your workload.

Last verified October 10, 2026. 22 shared benchmarks.

MiniMax-M2.5

MiniMax

38.1

Rank #181 Confirmed

Qwen3.5 27B

Alibaba (Qwen)

41.4

Rank #127 Confirmed

Summary

They share 22 benchmarks with published results for both. MiniMax-M2.5 scores higher in 2 categories and Qwen3.5 27B in 6 categories; 8 gaps are clear of the uncertainty.
The widest gap is in math, where Qwen3.5 27B leads 36.8 to 25.8.
The biggest single-benchmark swing is NYT Connections (extended): 16.8% for MiniMax-M2.5 and 47.9% for Qwen3.5 27B.
MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.30 / $2.40 for Qwen3.5 27B.
Qwen3.5 27B accepts more context: 262K tokens versus 205K.

Side by side

MiniMax-M2.5 and Qwen3.5 27B specifications
	MiniMax-M2.5	Qwen3.5 27B
Provider	MiniMax	Alibaba (Qwen)
Noometry Index	38.1	41.4
Released	2026-02-12	2026-02-23
Weights	Open	Open
Context window	205K	262K
Max output	131K	66K
Input $ / M tokens	$0.30	$0.30
Output $ / M tokens	$1.20	$2.40
Results tracked	33	28

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

Category by category

Coding MiniMax-M2.5 leads

MiniMax-M2.5: 47.9 (#53), Qwen3.5 27B: 38.6 (#159)

Coding benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena WebDev	1386	1357
LMArena Coding	1380	1427
ALE-Bench	618.17	349.45
SWE-bench Verified (bash only)	75.8%	—
SWE-bench Multilingual	68.3%	—
WeirdML	—	39.5%

Agentic & Tool Use Not comparable

MiniMax-M2.5: 30.2 (#70), Qwen3.5 27B: —

Agentic & Tool Use benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
Vending-Bench 2	-23.16	201.98
Terminal-Bench	42.7%	—

Reasoning Qwen3.5 27B leads

MiniMax-M2.5: 17.6 (#281), Qwen3.5 27B: 26.6 (#116)

Reasoning benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
NYT Connections (extended)	16.8%	47.9%
LMArena Hard Prompts	1372	1414
ARC-AGI-2	4.9%	—
Kagi LLM Benchmark	55.2%	—
ARC-AGI-1	63.7%	—
Thematic Generalization	—	45.5%
DTBench	—	82.4%
LMCA	—	34%
Epoch Capabilities Index	146.68	—

Math Qwen3.5 27B leads

MiniMax-M2.5: 25.8 (#242), Qwen3.5 27B: 36.8 (#103)

Math benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Math	1378	1428
MathArena Final-Answer Competitions	—	56.7%
ProofBench	4%	—

Knowledge MiniMax-M2.5 leads

MiniMax-M2.5: 38.6 (#132), Qwen3.5 27B: 37.4 (#149)

Knowledge benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
Vectara Hallucination Rate	9.1%	12.1%
LMArena Expert	1378	1428

Multimodal Not comparable

MiniMax-M2.5: —, Qwen3.5 27B: 38.7 (#56)

Multimodal benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Vision	—	1240

Multilingual Qwen3.5 27B leads

MiniMax-M2.5: 47.3 (#147), Qwen3.5 27B: 51.0 (#112)

Multilingual benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Non-English	1338	1390
LMArena Chinese	1392	1478
LMArena French	1363	1409
LMArena German	1361	1393
LMArena Japanese	1169	1345
LMArena Korean	1232	1359
LMArena Russian	1359	1389
LMArena Spanish	1354	1408

Instruction Following Qwen3.5 27B leads

MiniMax-M2.5: 71.7 (#143), Qwen3.5 27B: 73.7 (#114)

Instruction Following benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Instruction Following	1353	1393

Long Context Qwen3.5 27B leads

MiniMax-M2.5: 37.4 (#211), Qwen3.5 27B: 43.1 (#103)

Long Context benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Longer Query	1367	1413
CL-bench	11.4%	—
CL-bench Life	6.3%	—

Writing & Preference Qwen3.5 27B leads

MiniMax-M2.5: 54.1 (#149), Qwen3.5 27B: 59.5 (#106)

Writing & Preference benchmarks
Benchmark	MiniMax-M2.5	Qwen3.5 27B
LMArena Text	1359	1409
LMArena Creative Writing	1331	1362
LMArena Multi-Turn	1364	1410
EQ-Bench Creative Writing	1361	—

Frequently asked questions

Is MiniMax-M2.5 better than Qwen3.5 27B?

Which is cheaper, MiniMax-M2.5 or Qwen3.5 27B?

MiniMax-M2.5 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-M2.5 or Qwen3.5 27B better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 38.6 in the Noometry coding category.

Which has the bigger context window?

Qwen3.5 27B does, with 262K tokens against 205K.

How many benchmarks do MiniMax-M2.5 and Qwen3.5 27B share?

22 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Qwen3.5 27B has 28.