Compare AI models

Specifications
Parameters
1.3B/5.7B/6.7B/33B16B
Context window
16k8k
Benchmarks
HumanEval
79.3%48.1%
Overview
CompanyDeepSeekMoonshot AI
Release dateNov 2 2023Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder leads Moonlight on 1 of the 1 benchmark they both report (HumanEval). DeepSeek Coder shipped 478 days before Moonlight, so benchmark comparisons should account for the intervening progress.

DeepSeek Coder has 1.3B/5.7B/6.7B/33B parameters, while Moonlight has 16B. Context windows are 16k (DeepSeek Coder) vs 8k (Moonlight).

On HumanEval, DeepSeek Coder leads at 79.3% vs Moonlight at 48.1%.