Compare AI models

Specifications
Parameters
—16B
Context window
256k8k
Benchmarks
LiveCodeBench
37.9%14.3%
HumanEval
86.6%48.1%
Overview
CompanyMistralMoonshot AI
Release dateJan 13 2025Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Codestral 25.01 leads Moonlight on 2 of the 2 benchmarks they both report (LiveCodeBench, HumanEval). Codestral 25.01 shipped 40 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 256k (Codestral 25.01) vs 8k (Moonlight). Codestral 25.01 is closed, while Moonlight is open weight.

On LiveCodeBench, Codestral 25.01 leads at 37.9% vs Moonlight at 14.3%. On HumanEval, Codestral 25.01 leads at 86.6% vs Moonlight at 48.1%.