Compare AI models

Specifications
Parameters
—16B
Context window
2M8k
Benchmarks
LiveCodeBench
34.3%14.3%
MATH
86.5%45.3%
GPQA Diamond
59.1%35.2%
MMLU-Pro
75.8%42.4%
Overview
CompanyGoogleMoonshot AI
Release dateSep 24 2024Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Gemini 1.5 Pro-002 leads Moonlight on 4 of the 4 benchmarks they both report (LiveCodeBench, MATH, GPQA Diamond, MMLU-Pro). Gemini 1.5 Pro-002 shipped 151 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 2M (Gemini 1.5 Pro-002) vs 8k (Moonlight). Gemini 1.5 Pro-002 is closed, while Moonlight is open weight.

On LiveCodeBench, Gemini 1.5 Pro-002 leads at 34.3% vs Moonlight at 14.3%. On MATH, Gemini 1.5 Pro-002 leads at 86.5% vs Moonlight at 45.3%. On GPQA Diamond, Gemini 1.5 Pro-002 leads at 59.1% vs Moonlight at 35.2%. On MMLU-Pro, Gemini 1.5 Pro-002 leads at 75.8% vs Moonlight at 42.4%.