Compare AI models

Specifications
Parameters
24B9B
Context window
32k128k
API pricing
Cheapest input
$0.04DeepInfra—
Cheapest output
$0.064DeepInfra—
$14.00DeepInfra · fp8—
Benchmarks
HumanEval
84.8%82.3%
Overview
CompanyMistralZ.ai
Release dateJan 30 2025Jul 5 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Small 3 leads CodeGeeX4-ALL-9B on 1 of the 1 benchmark they both report (HumanEval). CodeGeeX4-ALL-9B shipped 209 days before Mistral Small 3, so benchmark comparisons should account for the intervening progress.

Mistral Small 3 has 24B parameters, while CodeGeeX4-ALL-9B has 9B. Context windows are 32k (Mistral Small 3) vs 128k (CodeGeeX4-ALL-9B).

On HumanEval, Mistral Small 3 leads at 84.8% vs CodeGeeX4-ALL-9B at 82.3%.