Model comparison

Devstral 2 vs Qwen3.6 Max Preview

Qwen3.6 Max Preview has enough public results to be ranked (#43); Devstral 2 does not yet, so treat this comparison as directional.

Last verified . 1 shared benchmarks.

Devstral 2 Mistral AI

36.1

Unranked Sparse

Qwen3.6 Max Preview Alibaba (Qwen)

51.5

Rank #43 Confirmed

Summary

  • They share 1 benchmark with published results for both. Devstral 2 scores higher in 0 categories and Qwen3.6 Max Preview in 1 category; one gap is clear of the uncertainty.
  • The widest gap is in coding, where Qwen3.6 Max Preview leads 48.7 to 33.2.
  • Devstral 2 is cheaper at $0.40 / $2 per million input/output tokens, against $1.30 / $7.80 for Qwen3.6 Max Preview.
  • Devstral 2 has downloadable open weights; the other is API-only.

Side by side

Devstral 2 and Qwen3.6 Max Preview specifications
Devstral 2Qwen3.6 Max Preview
ProviderMistral AIAlibaba (Qwen)
Noometry Index36.151.5
Released2025-12-092026-04-20
WeightsOpenProprietary
Context window262K262K
Max output262K66K
Input $ / M tokens$0.40$1.30
Output $ / M tokens$2$7.80
Results tracked229

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

Category by category

Coding Qwen3.6 Max Preview leads

Devstral 2: 33.2 (#261), Qwen3.6 Max Preview: 48.7 (#54)

Coding benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
LMArena WebDev11971482
SWE-bench Verified—76.7%
SWE-bench Verified (bash only)53.8%—
LMArena Coding—1471

Agentic & Tool Use Not comparable

Devstral 2: —, Qwen3.6 Max Preview: —

Agentic & Tool Use benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
Vending-Bench 2—4,254

Reasoning Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 41.7 (#53)

Reasoning benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
SimpleBench—63%
NYT Connections (extended)—74.1%
Chess Puzzles—20%
LMArena Hard Prompts—1457
Mystery Game Puzzles—19%
DTBench—87.2%
LMCA—42.5%
Epoch Capabilities Index—149.24

Math Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 54.1 (#46)

Math benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
OTIS Mock AIME 2024-2025—91.1%
LMArena Math—1465
FrontierMath (Feb 2025 set)—23.1%
FrontierMath Tier 4 (v1)—4.2%

Knowledge Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 57.6 (#39)

Knowledge benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
GPQA Diamond—87.4%
SimpleQA Verified—52%
LMArena Expert—1478

Multilingual Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 54.2 (#48)

Multilingual benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
LMArena Non-English—1437
LMArena Chinese—1487
LMArena French—1449
LMArena Russian—1445
LMArena Spanish—1454

Instruction Following Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 75.7 (#55)

Instruction Following benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
LMArena Instruction Following—1438

Long Context Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 44.6 (#61)

Long Context benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
LMArena Longer Query—1457

Writing & Preference Not comparable

Devstral 2: —, Qwen3.6 Max Preview: 63.8 (#60)

Writing & Preference benchmarks
BenchmarkDevstral 2Qwen3.6 Max Preview
LMArena Text—1447
LMArena Creative Writing—1435
LMArena Multi-Turn—1456

Frequently asked questions

Is Devstral 2 better than Qwen3.6 Max Preview?

Qwen3.6 Max Preview has enough public results to be ranked (#43); Devstral 2 does not yet, so treat this comparison as directional.

Which is cheaper, Devstral 2 or Qwen3.6 Max Preview?

Devstral 2 is cheaper. It lists at $0.40 per million input tokens and $2 per million output tokens; Qwen3.6 Max Preview lists at $1.30 and $7.80.

Is Devstral 2 or Qwen3.6 Max Preview better for coding?

Qwen3.6 Max Preview scores higher on coding benchmarks: 48.7 versus 33.2 in the Noometry coding category.

Which has the bigger context window?

Both accept 262K tokens.

How many benchmarks do Devstral 2 and Qwen3.6 Max Preview share?

1 benchmark has published results for both models. Devstral 2 has 2 scored results on Noometry and Qwen3.6 Max Preview has 29.

Related comparisons

Go deeper