Compare AI models

Specifications
Parameters
12B14.3B
Context window
128k32k
Benchmarks
HumanEval
72%34.2%
MMLU
69.2%62.5%
Overview
CompanyMistralQwen
Release dateSep 11 2024Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Pixtral 24.09 leads Qwen1.5-MoE on 2 of the 2 benchmarks they both report (HumanEval, MMLU). Qwen1.5-MoE shipped 167 days before Pixtral 24.09, so benchmark comparisons should account for the intervening progress.

Pixtral 24.09 has 12B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 128k (Pixtral 24.09) vs 32k (Qwen1.5-MoE).

On HumanEval, Pixtral 24.09 leads at 72% vs Qwen1.5-MoE at 34.2%. On MMLU, Pixtral 24.09 leads at 69.2% vs Qwen1.5-MoE at 62.5%.