Noometry

Model comparison

GPT-5.4 nano vs Qwen2-72B

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.0 on the Noometry Index.

Last verified . 20 shared benchmarks.

GPT-5.4 nano OpenAI

41.9

Rank #125 Confirmed

Qwen2-72B Alibaba (Qwen)

30.0

Rank #300 Confirmed

Summary

  • They share 20 benchmarks with published results for both. GPT-5.4 nano scores higher in 8 categories and Qwen2-72B in 0 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where GPT-5.4 nano leads 41.9 to 21.2.
  • The biggest single-benchmark swing is WeirdML: 49.2% for GPT-5.4 nano and 11.3% for Qwen2-72B.
  • Qwen2-72B has downloadable open weights; the other is API-only.

Side by side

GPT-5.4 nanoQwen2-72B
ProviderOpenAIAlibaba (Qwen)
Noometry Index41.930.0
Released2026-03-172024-06-07
WeightsProprietaryOpen
Context window400K—
Max output128K—
Input $ / M tokens$0.20—
Output $ / M tokens$1.25—
Results tracked4026

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

Category by category

Coding GPT-5.4 nano leads

GPT-5.4 nano: 43.6 (#84), Qwen2-72B: 29.1 (#310)

BenchmarkGPT-5.4 nanoQwen2-72B
WeirdML49.2%11.3%
LMArena Coding14051196
SciCode46.9%—
BigCodeBench Instruct—38.5%
BigCodeBench Complete—54%
ALE-Bench1,005—

Agentic & Tool Use Not comparable

GPT-5.4 nano: —, Qwen2-72B: 17.0 (#146)

BenchmarkGPT-5.4 nanoQwen2-72B
TheAgentCompany—1.1%
METR Time Horizons—29.9%

Reasoning Too close to call

GPT-5.4 nano: 23.7 (#173), Qwen2-72B: 23.2 (#181)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Hard Prompts13811191
Epoch Capabilities Index145.81125.28
ARC-AGI-25.7%—
Kagi LLM Benchmark39.7%—
ARC-AGI-151.5%—
CritPt9.3%—
Chess Puzzles30%—
Mystery Game Puzzles9%—
DTBench80.3%—
LMCA36.9%—
ForecastBench57.3—

Math GPT-5.4 nano leads

GPT-5.4 nano: 40.9 (#88), Qwen2-72B: 30.2 (#236)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Math14061235
FrontierMath (Tiers 1-3)44.9%—
FrontierMath Tier 412.2%—
OTIS Mock AIME 2024-202587.8%—
ProofBench5%—
MATH Level 5—39.1%
FrontierMath (Feb 2025 set)25.9%—
FrontierMath Tier 4 (v1)6.3%—

Knowledge GPT-5.4 nano leads

GPT-5.4 nano: 41.9 (#103), Qwen2-72B: 21.2 (#275)

BenchmarkGPT-5.4 nanoQwen2-72B
GPQA Diamond78.5%40.8%
LMArena Expert13961171
SimpleQA Verified11.7%—
Vectara Hallucination Rate3.1%—
MMLU—82.4%

Multimodal Not comparable

GPT-5.4 nano: 36.7 (#78), Qwen2-72B: —

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Vision1196—

Multilingual GPT-5.4 nano leads

GPT-5.4 nano: 48.6 (#140), Qwen2-72B: 35.9 (#244)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Non-English13591176
LMArena Chinese13921240
LMArena French13961170
LMArena German13671151
LMArena Japanese13431111
LMArena Korean13201083
LMArena Russian13631169
LMArena Spanish13711169

Instruction Following GPT-5.4 nano leads

GPT-5.4 nano: 71.9 (#144), Qwen2-72B: 61.7 (#241)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Instruction Following13621181

Long Context GPT-5.4 nano leads

GPT-5.4 nano: 41.6 (#137), Qwen2-72B: 36.1 (#235)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Longer Query13661192

Writing & Preference GPT-5.4 nano leads

GPT-5.4 nano: 55.7 (#142), Qwen2-72B: 40.8 (#241)

BenchmarkGPT-5.4 nanoQwen2-72B
LMArena Text13721203
LMArena Creative Writing13141181
LMArena Multi-Turn13821196

Frequently asked questions

Is GPT-5.4 nano better than Qwen2-72B?

GPT-5.4 nano is the stronger model overall, scoring 41.9 to 30.0 on the Noometry Index.

Is GPT-5.4 nano or Qwen2-72B better for coding?

GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 29.1 in the Noometry coding category.

How many benchmarks do GPT-5.4 nano and Qwen2-72B share?

20 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Qwen2-72B has 26.

Related comparisons

Go deeper