Noometry

Model comparison

Mistral Large 3 vs Qwen2.5-VL 72B Instruct

Mistral Large 3 is the stronger model overall, scoring 38.6 to 29.4 on the Noometry Index.

Last verified . 2 shared benchmarks.

Mistral Large 3 Mistral AI

38.6

Rank #172 Confirmed

Qwen2.5-VL 72B Instruct Alibaba (Qwen)

29.4

Rank #298 Confirmed

Summary

  • They share 2 benchmarks with published results for both. Mistral Large 3 scores higher in 1 category and Qwen2.5-VL 72B Instruct in 1 category; 2 gaps are clear of the uncertainty.
  • The widest gap is in reasoning, where Qwen2.5-VL 72B Instruct leads 20.0 to 14.6.
  • The biggest single-benchmark swing is Kagi LLM Benchmark: 50.9% for Mistral Large 3 and 36% for Qwen2.5-VL 72B Instruct.
  • Mistral Large 3 is cheaper at $0.25 / $0.75 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
  • Mistral Large 3 accepts more context: 262K tokens versus 131K.

Side by side

Mistral Large 3Qwen2.5-VL 72B Instruct
ProviderMistral AIAlibaba (Qwen)
Noometry Index38.629.4
Released2025-12-022024-09
WeightsOpenOpen
Context window262K131K
Max output8K8K
Input $ / M tokens$0.25$2.80
Output $ / M tokens$0.75$8.40
Results tracked246

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

Category by category

Coding Not comparable

Mistral Large 3: 34.1 (#230), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena WebDev1230—
LMArena Coding1447—

Agentic & Tool Use Not comparable

Mistral Large 3: —, Qwen2.5-VL 72B Instruct: 18.4 (#136)

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
OSWorld—5%

Reasoning Qwen2.5-VL 72B Instruct leads

Mistral Large 3: 14.6 (#314), Qwen2.5-VL 72B Instruct: 20.0 (#228)

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
Kagi LLM Benchmark50.9%36%
NYT Connections (extended)7.5%—
Thematic Generalization23%—
LMArena Hard Prompts1429—

Math Not comparable

Mistral Large 3: 36.0 (#131), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Math1415—

Knowledge Not comparable

Mistral Large 3: 35.5 (#173), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
Vectara Hallucination Rate14.5%—
LMArena Expert1421—

Multimodal Mistral Large 3 leads

Mistral Large 3: 37.6 (#65), Qwen2.5-VL 72B Instruct: 32.9 (#92)

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Vision12221107
Video-MME—73.5%
GeoBench—62%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Mistral Large 3: 52.6 (#79), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Non-English1412—
LMArena Chinese1448—
LMArena French1456—
LMArena German1438—
LMArena Japanese1399—
LMArena Korean1384—
LMArena Russian1412—
LMArena Spanish1437—

Instruction Following Not comparable

Mistral Large 3: 74.1 (#104), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Instruction Following1402—

Long Context Not comparable

Mistral Large 3: 43.1 (#102), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Longer Query1413—

Writing & Preference Not comparable

Mistral Large 3: 60.2 (#100), Qwen2.5-VL 72B Instruct: —

BenchmarkMistral Large 3Qwen2.5-VL 72B Instruct
LMArena Text1427—
LMArena Creative Writing1385—
EQ-Bench Creative Writing1412—
LMArena Multi-Turn1429—

Frequently asked questions

Is Mistral Large 3 better than Qwen2.5-VL 72B Instruct?

Mistral Large 3 is the stronger model overall, scoring 38.6 to 29.4 on the Noometry Index.

Which is cheaper, Mistral Large 3 or Qwen2.5-VL 72B Instruct?

Mistral Large 3 is cheaper. It lists at $0.25 per million input tokens and $0.75 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

Which has the bigger context window?

Mistral Large 3 does, with 262K tokens against 131K.

How many benchmarks do Mistral Large 3 and Qwen2.5-VL 72B Instruct share?

2 benchmarks have published results for both models. Mistral Large 3 has 24 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper