Compare AI models

Specifications
Parameters
—16B
Context window
200k8k
Benchmarks
HumanEval
84.9%48.1%
GPQA Diamond
50.4%35.2%
MMLU
88.2%70%
GSM8K
95%77.4%
Overview
CompanyAnthropicMoonshot AI
Release dateMar 4 2024Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude 3 Opus leads Moonlight on 4 of the 4 benchmarks they both report (HumanEval, GPQA Diamond, MMLU, GSM8K). Claude 3 Opus shipped 355 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude 3 Opus) vs 8k (Moonlight). Claude 3 Opus is closed, while Moonlight is open weight.

On HumanEval, Claude 3 Opus leads at 84.9% vs Moonlight at 48.1%. On GPQA Diamond, Claude 3 Opus leads at 50.4% vs Moonlight at 35.2%. On MMLU, Claude 3 Opus leads at 88.2% vs Moonlight at 70%. On GSM8K, Claude 3 Opus leads at 95% vs Moonlight at 77.4%.