Model comparison

Devstral 2 vs Qwen3.7 Max

Qwen3.7 Max has enough public results to be ranked (#35); Devstral 2 does not yet, so treat this comparison as directional.

Last verified October 10, 2026. 1 shared benchmarks.

Devstral 2

Mistral AI

35.8

Unranked Sparse

Qwen3.7 Max

Alibaba (Qwen)

51.4

Rank #35 Confirmed

Summary

They share 1 benchmark with published results for both. Devstral 2 scores higher in 0 categories and Qwen3.7 Max in 1 category; one gap is clear of the uncertainty.
The widest gap is in coding, where Qwen3.7 Max leads 50.2 to 33.0.
Devstral 2 is cheaper at $0.40 / $2 per million input/output tokens, against $2.50 / $7.50 for Qwen3.7 Max.
Qwen3.7 Max accepts more context: 1M tokens versus 262K.
Devstral 2 has downloadable open weights; the other is API-only.

Side by side

Devstral 2 and Qwen3.7 Max specifications
	Devstral 2	Qwen3.7 Max
Provider	Mistral AI	Alibaba (Qwen)
Noometry Index	35.8	51.4
Released	2025-12-09	2026-05-19
Weights	Open	Proprietary
Context window	262K	1M
Max output	262K	131K
Input $ / M tokens	$0.40	$2.50
Output $ / M tokens	$2	$7.50
Results tracked	2	33

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

Category by category

Coding Qwen3.7 Max leads

Devstral 2: 33.0 (#253), Qwen3.7 Max: 50.2 (#43)

Coding benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
LMArena WebDev	1196	1515
SWE-bench Verified	—	77.3%
SWE-bench Verified (bash only)	53.8%	—
SciCode	—	48.8%
LMArena Coding	—	1498
ALE-Bench	—	1,189

Agentic & Tool Use Not comparable

Devstral 2: —, Qwen3.7 Max: 21.8 (#129)

Agentic & Tool Use benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
GBAEval	—	0.4%

Reasoning Not comparable

Devstral 2: —, Qwen3.7 Max: 48.4 (#36)

Reasoning benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
SimpleBench	—	70.4%
NYT Connections (extended)	—	85.1%
CritPt	—	13.4%
Chess Puzzles	—	19%
EBR-Bench	—	9.5%
LMArena Hard Prompts	—	1483
Mystery Game Puzzles	—	32%
DTBench	—	92.3%
LMCA	—	44%
Epoch Capabilities Index	—	153.74

Math Not comparable

Devstral 2: —, Qwen3.7 Max: 62.3 (#26)

Math benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
FrontierMath (Tiers 1-3)	—	64.6%
FrontierMath Tier 4	—	34.1%
OTIS Mock AIME 2024-2025	—	95.6%
ProofBench	—	26%
LMArena Math	—	1488

Knowledge Not comparable

Devstral 2: —, Qwen3.7 Max: 61.7 (#24)

Knowledge benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
GPQA Diamond	—	90.9%
SimpleQA Verified	—	55.8%
LMArena Expert	—	1488

Multilingual Not comparable

Devstral 2: —, Qwen3.7 Max: 57.1 (#13)

Multilingual benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
LMArena Non-English	—	1475
LMArena Chinese	—	1533
LMArena Russian	—	1482

Instruction Following Not comparable

Devstral 2: —, Qwen3.7 Max: 76.9 (#34)

Instruction Following benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
LMArena Instruction Following	—	1460

Long Context Not comparable

Devstral 2: —, Qwen3.7 Max: 45.4 (#34)

Long Context benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
LMArena Longer Query	—	1484

Writing & Preference Not comparable

Devstral 2: —, Qwen3.7 Max: 65.2 (#52)

Writing & Preference benchmarks
Benchmark	Devstral 2	Qwen3.7 Max
LMArena Text	—	1476
LMArena Creative Writing	—	1450
EQ-Bench 4	—	1110
LMArena Multi-Turn	—	1482

Frequently asked questions

Is Devstral 2 better than Qwen3.7 Max?

Qwen3.7 Max has enough public results to be ranked (#35); Devstral 2 does not yet, so treat this comparison as directional.

Which is cheaper, Devstral 2 or Qwen3.7 Max?

Devstral 2 is cheaper. It lists at $0.40 per million input tokens and $2 per million output tokens; Qwen3.7 Max lists at $2.50 and $7.50.

Is Devstral 2 or Qwen3.7 Max better for coding?

Qwen3.7 Max scores higher on coding benchmarks: 50.2 versus 33.0 in the Noometry coding category.

Which has the bigger context window?

Qwen3.7 Max does, with 1M tokens against 262K.

How many benchmarks do Devstral 2 and Qwen3.7 Max share?

1 benchmark has published results for both models. Devstral 2 has 2 scored results on Noometry and Qwen3.7 Max has 33.