Compare AI models

Specifications
Parameters
24B7B
Context window
32k64k
API pricing
Cheapest input
$0.04DeepInfra—
Cheapest output
$0.064DeepInfra—
$14.00DeepInfra · fp8—
Benchmarks
HumanEval
84.8%83.5%
Overview
CompanyMistralQwen
Release dateJan 30 2025Apr 16 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Small 3 leads CodeQwen1.5 on 1 of the 1 benchmark they both report (HumanEval). CodeQwen1.5 shipped 289 days before Mistral Small 3, so benchmark comparisons should account for the intervening progress.

Mistral Small 3 has 24B parameters, while CodeQwen1.5 has 7B. Context windows are 32k (Mistral Small 3) vs 64k (CodeQwen1.5).

On HumanEval, Mistral Small 3 leads at 84.8% vs CodeQwen1.5 at 83.5%.