Noometry

Model comparison

Qwen Max vs Qwen2.5-VL 72B Instruct

Qwen Max is the stronger model overall, scoring 34.4 to 29.4 on the Noometry Index.

Last verified . 0 shared benchmarks.

Qwen Max Alibaba (Qwen)

34.4

Rank #226 Confirmed

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

29.4

Rank #298 Confirmed

Summary

  • The widest gap is in reasoning, where Qwen Max leads 24.1 to 20.0.
  • Qwen Max is cheaper at $1.60 / $6.40 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
  • Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 33K.
  • Qwen2.5-VL 72B Instruct has downloadable open weights; the other is API-only.

Side by side

Qwen MaxQwen2.5-VL 72B Instruct
ProviderAlibaba (Qwen)Alibaba (Qwen)
Noometry Index34.429.4
Released2024-04-032024-09
WeightsProprietaryOpen
Context window33K131K
Max output8K8K
Input $ / M tokens$1.60$2.80
Output $ / M tokens$6.40$8.40
Results tracked236

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

Category by category

Coding Not comparable

Qwen Max: 30.3 (#288), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
Aider Polyglot21.8%—
LMArena Coding1288—

Agentic & Tool Use Not comparable

Qwen Max: —, Qwen2.5-VL 72B Instruct: 18.4 (#136)

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
OSWorld—5%

Reasoning Qwen Max leads

Qwen Max: 24.1 (#154), Qwen2.5-VL 72B Instruct: 20.0 (#228)

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
Kagi LLM Benchmark—36%
LMArena Hard Prompts1270—

Math Not comparable

Qwen Max: 21.0 (#272), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
OTIS Mock AIME 2024-202516.1%—
LMArena Math1275—
MATH Level 567.2%—
FrontierMath (Feb 2025 set)1%—

Knowledge Not comparable

Qwen Max: 30.3 (#222), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
GPQA Diamond56.1%—
LMArena Expert1248—

Multimodal Not comparable

Qwen Max: —, Qwen2.5-VL 72B Instruct: 32.9 (#92)

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
LMArena Vision—1107
Video-MME—73.5%
GeoBench—62%
SpatialViz-Bench—33.3%

Multilingual Not comparable

Qwen Max: 42.0 (#197), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
LMArena Non-English1263—
LMArena Chinese1254—
LMArena French1329—
LMArena German1254—
LMArena Japanese1205—
LMArena Korean1141—
LMArena Russian1274—
LMArena Spanish1290—

Instruction Following Not comparable

Qwen Max: 66.7 (#203), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
LMArena Instruction Following1262—

Long Context Not comparable

Qwen Max: 39.4 (#175), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
Fiction.LiveBench66.7%—
LMArena Longer Query1289—

Writing & Preference Not comparable

Qwen Max: 48.0 (#200), Qwen2.5-VL 72B Instruct: —

BenchmarkQwen MaxQwen2.5-VL 72B Instruct
LMArena Text1282—
LMArena Creative Writing1248—
LMArena Multi-Turn1277—

Frequently asked questions

Is Qwen Max better than Qwen2.5-VL 72B Instruct?

Qwen Max is the stronger model overall, scoring 34.4 to 29.4 on the Noometry Index.

Which is cheaper, Qwen Max or Qwen2.5-VL 72B Instruct?

Qwen Max is cheaper. It lists at $1.60 per million input tokens and $6.40 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

Which has the bigger context window?

Qwen2.5-VL 72B Instruct does, with 131K tokens against 33K.

How many benchmarks do Qwen Max and Qwen2.5-VL 72B Instruct share?

0 benchmarks have published results for both models. Qwen Max has 23 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.

Related comparisons

Go deeper