Compare AI models

Specifications
Parameters
30B16B
Context window
128k8k
API pricing
Cheapest input
$0.30DeepInfra—
Cheapest output
$1.20DeepInfra—
$120ModelRun · bf16—
Benchmarks
GPQA Diamond
83.5%35.2%
Overview
CompanyMetaMoonshot AI
Release dateAug 10 2026Feb 22 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Muse Glimmer leads Moonlight on 1 of the 1 benchmark they both report (GPQA Diamond). Moonlight shipped 534 days before Muse Glimmer, so benchmark comparisons should account for the intervening progress.

Muse Glimmer has 30B parameters, while Moonlight has 16B. Context windows are 128k (Muse Glimmer) vs 8k (Moonlight).

On GPQA Diamond, Muse Glimmer leads at 83.5% vs Moonlight at 35.2%.