Compare AI models

Specifications
Parameters
7B16B
Context window
256k8k
Benchmarks
HumanEval
75%48.1%
Overview
CompanyMistralMoonshot AI
Release dateJul 16 2024Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Codestral Mamba 7B leads Moonlight on 1 of the 1 benchmark they both report (HumanEval). Codestral Mamba 7B shipped 221 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Codestral Mamba 7B has 7B parameters, while Moonlight has 16B. Context windows are 256k (Codestral Mamba 7B) vs 8k (Moonlight).

On HumanEval, Codestral Mamba 7B leads at 75% vs Moonlight at 48.1%.