Compare AI models

Specifications
Parameters
12B14B
Context window
128k2k
Benchmarks
HumanEval
72%32.3%
MATH
48.1%24.8%
MMLU
69.2%66.3%
Overview
CompanyMistralQwen
Release dateSep 11 2024Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Pixtral 24.09 leads Qwen-14B on 3 of the 3 benchmarks they both report (HumanEval, MATH, MMLU). Qwen-14B shipped 352 days before Pixtral 24.09, so benchmark comparisons should account for the intervening progress.

Pixtral 24.09 has 12B parameters, while Qwen-14B has 14B. Context windows are 128k (Pixtral 24.09) vs 2k (Qwen-14B).

On HumanEval, Pixtral 24.09 leads at 72% vs Qwen-14B at 32.3%. On MATH, Pixtral 24.09 leads at 48.1% vs Qwen-14B at 24.8%. On MMLU, Pixtral 24.09 leads at 69.2% vs Qwen-14B at 66.3%.