Compare AI models

Specifications
Parameters
16B—
Context window
8k1M
API pricing
Cheapest input
—$0.10Azure
Cheapest output
—$0.40Azure
—$40.00Azure
Benchmarks
HumanEval
48.1%87%
GPQA Diamond
35.2%50.3%
SimpleQA
5.6%7.6%
MMLU
70%80.1%
Overview
CompanyMoonshot AIOpenAI
Release dateFeb 22 2025Apr 14 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GPT-4.1 nano leads Moonlight on 4 of the 4 benchmarks they both report (HumanEval, GPQA Diamond, SimpleQA, MMLU). Moonlight shipped 51 days before GPT-4.1 nano, so benchmark comparisons should account for the intervening progress.

Context windows are 8k (Moonlight) vs 1M (GPT-4.1 nano). Moonlight is open weight, while GPT-4.1 nano is closed.

On HumanEval, GPT-4.1 nano leads at 87% vs Moonlight at 48.1%. On GPQA Diamond, GPT-4.1 nano leads at 50.3% vs Moonlight at 35.2%. On SimpleQA, GPT-4.1 nano leads at 7.6% vs Moonlight at 5.6%. On MMLU, GPT-4.1 nano leads at 80.1% vs Moonlight at 70%.