Compare AI models

Specifications
Parameters
—14B
Context window
100k2k
Benchmarks
HumanEval
71.2%32.3%
MMLU
78.5%66.3%
GSM8K
88%61.3%
Overview
CompanyAnthropicQwen
Release dateJul 11 2023Sep 25 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude 2 leads Qwen-14B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Claude 2 shipped 76 days before Qwen-14B, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude 2) vs 2k (Qwen-14B). Claude 2 is closed, while Qwen-14B is open weight.

On HumanEval, Claude 2 leads at 71.2% vs Qwen-14B at 32.3%. On MMLU, Claude 2 leads at 78.5% vs Qwen-14B at 66.3%. On GSM8K, Claude 2 leads at 88% vs Qwen-14B at 61.3%.