Compare AI models

Specifications
Parameters
22B110B
Context window
32k32k
Benchmarks
HumanEval
81.1%52.4%
Overview
CompanyMistralQwen
Release dateMay 29 2024Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Codestral 22B leads Qwen1.5-110B on 1 of the 1 benchmark they both report (HumanEval). Qwen1.5-110B shipped 34 days before Codestral 22B, so benchmark comparisons should account for the intervening progress.

Codestral 22B has 22B parameters, while Qwen1.5-110B has 110B. Context windows are 32k (Codestral 22B) vs 32k (Qwen1.5-110B).

On HumanEval, Codestral 22B leads at 81.1% vs Qwen1.5-110B at 52.4%.