Compare AI models

Specifications
Parameters
—14.3B
Context window
32k32k
Benchmarks
HumanEval
67.7%34.2%
MMLU
71.8%62.5%
GSM8K
77.9%61.5%
Overview
CompanyGoogleQwen
Release dateDec 13 2023Mar 28 2024
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Gemini 1.0 Pro leads Qwen1.5-MoE on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Gemini 1.0 Pro shipped 106 days before Qwen1.5-MoE, so benchmark comparisons should account for the intervening progress.

Context windows are 32k (Gemini 1.0 Pro) vs 32k (Qwen1.5-MoE). Gemini 1.0 Pro is closed, while Qwen1.5-MoE is open weight.

On HumanEval, Gemini 1.0 Pro leads at 67.7% vs Qwen1.5-MoE at 34.2%. On MMLU, Gemini 1.0 Pro leads at 71.8% vs Qwen1.5-MoE at 62.5%. On GSM8K, Gemini 1.0 Pro leads at 77.9% vs Qwen1.5-MoE at 61.5%.