Compare AI models

Specifications
Parameters
400B16B
Context window
1M8k
API pricing
Cheapest input
$0.35Google—
Cheapest output
$1.00Parasail—
$120Parasail · fp8—
Benchmarks
LiveCodeBench
43.4%14.3%
GPQA Diamond
69.8%35.2%
MMLU-Pro
80.5%42.4%
Overview
CompanyMetaMoonshot AI
Release dateApr 5 2025Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

LLaMA 4 Maverick leads Moonlight on 3 of the 3 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU-Pro). Moonlight shipped 42 days before LLaMA 4 Maverick, so benchmark comparisons should account for the intervening progress.

LLaMA 4 Maverick has 400B parameters, while Moonlight has 16B. Context windows are 1M (LLaMA 4 Maverick) vs 8k (Moonlight).

On LiveCodeBench, LLaMA 4 Maverick leads at 43.4% vs Moonlight at 14.3%. On GPQA Diamond, LLaMA 4 Maverick leads at 69.8% vs Moonlight at 35.2%. On MMLU-Pro, LLaMA 4 Maverick leads at 80.5% vs Moonlight at 42.4%.