Compare AI models

Specifications
Parameters
—0.6B/1.7B/4B/8B/14B/30B/32B/235B
Context window
128k—
API pricing
Cheapest input
—$0.064DeepInfra
Cheapest output
—$0.192DeepInfra
—$30.00DeepInfra · fp8
Benchmarks
MATH-500
85.5%98%
AIME 2024
44.6%85.7%
GPQA Diamond
73.3%71.1%
Overview
CompanyOpenAIQwen
Release dateSep 12 2024Apr 29 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen3 leads o1-preview on 2 of the 3 benchmarks they both report (MATH-500, AIME 2024, GPQA Diamond). o1-preview shipped 229 days before Qwen3, so benchmark comparisons should account for the intervening progress.

o1-preview is closed, while Qwen3 is open weight.

On MATH-500, Qwen3 leads at 98% vs o1-preview at 85.5%. On AIME 2024, Qwen3 leads at 85.7% vs o1-preview at 44.6%. On GPQA Diamond, o1-preview leads at 73.3% vs Qwen3 at 71.1%.