Compare AI models

Specifications
Parameters
24B14B
Context window
32k2k
API pricing
Cheapest input
$0.04DeepInfra—
Cheapest output
$0.064DeepInfra—
$14.00DeepInfra · fp8—
Benchmarks
HumanEval
84.8%32.3%
MATH
70.6%24.8%
Overview
CompanyMistralQwen
Release dateJan 30 2025Sep 25 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Small 3 leads Qwen-14B on 2 of the 2 benchmarks they both report (HumanEval, MATH). Qwen-14B shipped 493 days before Mistral Small 3, so benchmark comparisons should account for the intervening progress.

Mistral Small 3 has 24B parameters, while Qwen-14B has 14B. Context windows are 32k (Mistral Small 3) vs 2k (Qwen-14B).

On HumanEval, Mistral Small 3 leads at 84.8% vs Qwen-14B at 32.3%. On MATH, Mistral Small 3 leads at 70.6% vs Qwen-14B at 24.8%.