Noometry

Model comparison

Phi-4 vs Qwen2.5-Max

Qwen2.5-Max is the stronger model overall, scoring 40.7 to 31.2 on the Noometry Index.

Last verified . 27 shared benchmarks.

Phi-4 Microsoft

31.2

Rank #279 Confirmed

Qwen2.5-Max Alibaba (Qwen)

40.7

Rank #146 Confirmed

Summary

  • They share 27 benchmarks with published results for both. Phi-4 scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen2.5-Max leads 36.9 to 20.8.
  • The biggest single-benchmark swing is LiveBench Coding: 30.7% for Phi-4 and 64.4% for Qwen2.5-Max.
  • Phi-4 has downloadable open weights; the other is API-only.

Side by side

Phi-4Qwen2.5-Max
ProviderMicrosoftAlibaba (Qwen)
Noometry Index31.240.7
Released2024-12-112025-01-25
WeightsOpenProprietary
Context window128K—
Max output4K—
Input $ / M tokens$0.07—
Output $ / M tokens$0.14—
Results tracked3727

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

Category by category

Coding Qwen2.5-Max leads

Phi-4: 34.4 (#239), Qwen2.5-Max: 41.8 (#117)

BenchmarkPhi-4Qwen2.5-Max
LiveBench Coding30.7%64.4%
LMArena Coding12311359
BigCodeBench Instruct45.5%—
BigCodeBench Complete55.4%—

Agentic & Tool Use Not comparable

Phi-4: 22.8 (#128), Qwen2.5-Max: —

BenchmarkPhi-4Qwen2.5-Max
Berkeley Function Calling Leaderboard28.8%—
BALROG11.6%—

Reasoning Qwen2.5-Max leads

Phi-4: 17.7 (#291), Qwen2.5-Max: 25.6 (#147)

BenchmarkPhi-4Qwen2.5-Max
LiveBench Reasoning47.8%51.4%
LMArena Hard Prompts12201360
LiveBench Data Analysis45.2%67.9%
Epoch Capabilities Index130.42132.53
LiveBench41.6%62.3%
Chess Puzzles1%—

Math Qwen2.5-Max leads

Phi-4: 20.8 (#285), Qwen2.5-Max: 36.9 (#162)

BenchmarkPhi-4Qwen2.5-Max
LiveBench Math42%58.4%
LMArena Math12461369
OTIS Mock AIME 2024-202513.8%—
MATH Level 564.9%—

Knowledge Qwen2.5-Max leads

Phi-4: 32.6 (#209), Qwen2.5-Max: 35.3 (#186)

BenchmarkPhi-4Qwen2.5-Max
Confabulations29.4%21.8%
LMArena Expert12031337
GPQA Diamond56.1%—
Vectara Hallucination Rate3.7%—
MMLU84.8%—

Multilingual Qwen2.5-Max leads

Phi-4: 37.2 (#237), Qwen2.5-Max: 48.1 (#146)

BenchmarkPhi-4Qwen2.5-Max
LMArena Non-English11971352
LMArena Chinese12121382
LMArena French12241396
LMArena German12221350
LMArena Japanese11581300
LMArena Korean11511304
LMArena Russian12091353
LMArena Spanish12341377

Instruction Following Qwen2.5-Max leads

Phi-4: 60.4 (#251), Qwen2.5-Max: 71.3 (#152)

BenchmarkPhi-4Qwen2.5-Max
LiveBench Instruction Following58.4%75.3%
LMArena Instruction Following12011335

Long Context Qwen2.5-Max leads

Phi-4: 36.9 (#226), Qwen2.5-Max: 41.4 (#142)

BenchmarkPhi-4Qwen2.5-Max
LMArena Longer Query12171358

Writing & Preference Qwen2.5-Max leads

Phi-4: 40.5 (#244), Qwen2.5-Max: 55.4 (#146)

BenchmarkPhi-4Qwen2.5-Max
LMArena Text12171367
LMArena Creative Writing11821339
Short-Story Creative Writing62.6%72.9%
LMArena Multi-Turn12061364
LiveBench Language25.6%56.3%

Frequently asked questions

Is Phi-4 better than Qwen2.5-Max?

Qwen2.5-Max is the stronger model overall, scoring 40.7 to 31.2 on the Noometry Index.

Is Phi-4 or Qwen2.5-Max better for coding?

Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 34.4 in the Noometry coding category.

How many benchmarks do Phi-4 and Qwen2.5-Max share?

27 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen2.5-Max has 27.

Related comparisons

Go deeper