Compare AI models

Specifications
Parameters
0.5B/1.5B/7B/57B/72B—
Context window
128k1M
Benchmarks
HumanEval
86%86.6%
MATH
69%81.1%
MMLU-Pro
64.4%64.5%
GSM8K
91.1%93.8%
Overview
CompanyQwenQwen
Release dateJun 6 2024Nov 15 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Turbo (1M) leads Qwen2 on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU-Pro, GSM8K). Qwen2 shipped 162 days before Qwen2.5-Turbo (1M), so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Qwen2) vs 1M (Qwen2.5-Turbo (1M)). Qwen2 is open weight, while Qwen2.5-Turbo (1M) is closed.

On HumanEval, Qwen2.5-Turbo (1M) leads at 86.6% vs Qwen2 at 86%. On MATH, Qwen2.5-Turbo (1M) leads at 81.1% vs Qwen2 at 69%. On MMLU-Pro, Qwen2.5-Turbo (1M) leads at 64.5% vs Qwen2 at 64.4%. On GSM8K, Qwen2.5-Turbo (1M) leads at 93.8% vs Qwen2 at 91.1%.