Compare · ModelsLive · 2 picked · head to head
o3 vs Qwen2-72B
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
o3 wins on 4/4 benchmarks
o3 wins 4 of 4 shared benchmarks. Leads in knowledge · math · general.
Category leads
knowledge·o3math·o3general·o3coding·o3
Hype vs Reality
Attention vs performance
o3
#121 by perf·no signal
Qwen2-72B
#216 by perf·#1 by attention
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
4 benchmarks · 2 models
o3Qwen2-72B
GPQA diamond
o3 leads by +53.8
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
o3
75.8
Qwen2-72B
21.9
MATH level 5
o3 leads by +58.7
MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics.
o3
97.8
Qwen2-72B
39.1
Metr Time Horizons
o3 leads by +35.5
o3
65.4
Qwen2-72B
29.9
WeirdML
o3 leads by +41.1
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
o3
52.4
Qwen2-72B
11.3
Full benchmark table
| Benchmark | o3 | Qwen2-72B |
|---|---|---|
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 75.8 | 21.9 |
MATH level 5 MATH Level 5 · the hardest tier of the MATH benchmark, featuring competition-level problems from AMC, AIME, and Olympiad-style mathematics. | 97.8 | 39.1 |
Metr Time Horizons | 65.4 | 29.9 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 52.4 | 11.3 |
Pricing · per 1M tokens · projected $/mo at 10M tokens