Compare AI models

Specifications
Parameters
—0.5B/1.5B/7B/57B/72B
Context window
128k128k
Benchmarks
HumanEval
92.4%86%
MMLU
90.8%82.3%
Overview
CompanyOpenAIQwen
Release dateSep 12 2024Jun 6 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

o1-preview leads Qwen2 on 2 of the 2 benchmarks they both report (HumanEval, MMLU). Qwen2 shipped 98 days before o1-preview, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (o1-preview) vs 128k (Qwen2). o1-preview is closed, while Qwen2 is open weight.

On HumanEval, o1-preview leads at 92.4% vs Qwen2 at 86%. On MMLU, o1-preview leads at 90.8% vs Qwen2 at 82.3%.