Compare AI models

Specifications
Parameters
7B/13B/70B7B/14B
Context window
4k1M
Benchmarks
HumanEval
29.9%88.4%
MATH
13.5%79.5%
GSM8K
56.8%94.8%
Overview
CompanyMetaQwen
Release dateJul 18 2023Jan 27 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-1M leads LLaMA 2 on 3 of the 3 benchmarks they both report (HumanEval, MATH, GSM8K). LLaMA 2 shipped 559 days before Qwen2.5-1M, so benchmark comparisons should account for the intervening progress.

LLaMA 2 has 7B/13B/70B parameters, while Qwen2.5-1M has 7B/14B. Context windows are 4k (LLaMA 2) vs 1M (Qwen2.5-1M).

On HumanEval, Qwen2.5-1M leads at 88.4% vs LLaMA 2 at 29.9%. On MATH, Qwen2.5-1M leads at 79.5% vs LLaMA 2 at 13.5%. On GSM8K, Qwen2.5-1M leads at 94.8% vs LLaMA 2 at 56.8%.