Compare AI models

Specifications
Parameters
—14.3B
Context window
1M32k
Benchmarks
HumanEval
74.3%34.2%
MMLU
78.9%62.5%
GSM8K
86.2%61.5%
Overview
CompanyGoogleQwen
Release dateMay 14 2024Mar 28 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Gemini 1.5 Flash leads Qwen1.5-MoE on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Qwen1.5-MoE shipped 47 days before Gemini 1.5 Flash, so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Gemini 1.5 Flash) vs 32k (Qwen1.5-MoE). Gemini 1.5 Flash is closed, while Qwen1.5-MoE is open weight.

On HumanEval, Gemini 1.5 Flash leads at 74.3% vs Qwen1.5-MoE at 34.2%. On MMLU, Gemini 1.5 Flash leads at 78.9% vs Qwen1.5-MoE at 62.5%. On GSM8K, Gemini 1.5 Flash leads at 86.2% vs Qwen1.5-MoE at 61.5%.