Compare AI models

Specifications
Parameters
2B/7B110B
Context window
8k32k
Benchmarks
HumanEval
32.3%52.4%
MATH
24.3%49.6%
HellaSwag
81.2%87.5%
MMLU
64.3%80.4%
GSM8K
46.4%85.4%
Overview
CompanyGoogleQwen
Release dateFeb 21 2024Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-110B leads Gemma 1 on 5 of the 5 benchmarks they both report. Gemma 1 shipped 64 days before Qwen1.5-110B, so benchmark comparisons should account for the intervening progress.

Gemma 1 has 2B/7B parameters, while Qwen1.5-110B has 110B. Context windows are 8k (Gemma 1) vs 32k (Qwen1.5-110B).

On HumanEval, Qwen1.5-110B leads at 52.4% vs Gemma 1 at 32.3%. On MATH, Qwen1.5-110B leads at 49.6% vs Gemma 1 at 24.3%. On HellaSwag, Qwen1.5-110B leads at 87.5% vs Gemma 1 at 81.2%. On MMLU, Qwen1.5-110B leads at 80.4% vs Gemma 1 at 64.3%. On GSM8K, Qwen1.5-110B leads at 85.4% vs Gemma 1 at 46.4%.