Compare AI models

Specifications
Parameters
1B/4B/12B/27B14B
Context window
—2k
API pricing
Cheapest input
$0.04DeepInfra—
Cheapest output
$0.05Parasail—
$10.50Parasail · bf16—
Benchmarks
HumanEval
87.8%32.3%
MATH
89%24.8%
MMLU
76.9%66.3%
GSM8K
95.9%61.3%
Overview
CompanyGoogleQwen
Release dateMar 12 2025Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Gemma 3 leads Qwen-14B on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Qwen-14B shipped 534 days before Gemma 3, so benchmark comparisons should account for the intervening progress.

Gemma 3 has 1B/4B/12B/27B parameters, while Qwen-14B has 14B.

On HumanEval, Gemma 3 leads at 87.8% vs Qwen-14B at 32.3%. On MATH, Gemma 3 leads at 89% vs Qwen-14B at 24.8%. On MMLU, Gemma 3 leads at 76.9% vs Qwen-14B at 66.3%. On GSM8K, Gemma 3 leads at 95.9% vs Qwen-14B at 61.3%.