Compare AI models

Specifications
Parameters
8B/70B7B/14B
Context window
8k1M
Benchmarks
HumanEval
81.7%88.4%
MATH
50.4%79.5%
GSM8K
93%94.8%
Overview
CompanyMetaQwen
Release dateApr 18 2024Jan 27 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-1M leads LLaMA 3 (8B/70B) on 3 of the 3 benchmarks they both report (HumanEval, MATH, GSM8K). LLaMA 3 (8B/70B) shipped 284 days before Qwen2.5-1M, so benchmark comparisons should account for the intervening progress.

LLaMA 3 (8B/70B) has 8B/70B parameters, while Qwen2.5-1M has 7B/14B. Context windows are 8k (LLaMA 3 (8B/70B)) vs 1M (Qwen2.5-1M).

On HumanEval, Qwen2.5-1M leads at 88.4% vs LLaMA 3 (8B/70B) at 81.7%. On MATH, Qwen2.5-1M leads at 79.5% vs LLaMA 3 (8B/70B) at 50.4%. On GSM8K, Qwen2.5-1M leads at 94.8% vs LLaMA 3 (8B/70B) at 93%.