Compare AI models

Specifications
Parameters
16B16B
Context window
32k8k
Benchmarks
HumanEval
57.3%48.1%
MATH
27.9%45.3%
MMLU
55.7%70%
GSM8K
72%77.4%
Overview
CompanyDeepSeekMoonshot AI
Release dateMay 16 2024Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Moonlight leads DeepSeek V2 Lite on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). DeepSeek V2 Lite shipped 282 days before Moonlight, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Moonlight has 16B. Context windows are 32k (DeepSeek V2 Lite) vs 8k (Moonlight).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Moonlight at 48.1%. On MATH, Moonlight leads at 45.3% vs DeepSeek V2 Lite at 27.9%. On MMLU, Moonlight leads at 70% vs DeepSeek V2 Lite at 55.7%. On GSM8K, Moonlight leads at 77.4% vs DeepSeek V2 Lite at 72%.