Compare AI models

Specifications
Parameters
—14.3B
Context window
100k32k
Benchmarks
HumanEval
71.2%34.2%
MMLU
78.5%62.5%
GSM8K
88%61.5%
Overview
CompanyAnthropicQwen
Release dateJul 11 2023Mar 28 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude 2 leads Qwen1.5-MoE on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Claude 2 shipped 261 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude 2) vs 32k (Qwen1.5-MoE). Claude 2 is closed, while Qwen1.5-MoE is open weight.

On HumanEval, Claude 2 leads at 71.2% vs Qwen1.5-MoE at 34.2%. On MMLU, Claude 2 leads at 78.5% vs Qwen1.5-MoE at 62.5%. On GSM8K, Claude 2 leads at 88% vs Qwen1.5-MoE at 61.5%.