Compare AI models

Specifications
Parameters
—16B
Context window
100k8k
Benchmarks
HumanEval
58.7%48.1%
GSM8K
86.7%77.4%
Overview
CompanyAnthropicMoonshot AI
Release dateAug 9 2023Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude Instant 1.2 leads Moonlight on 2 of the 2 benchmarks they both report (HumanEval, GSM8K). Claude Instant 1.2 shipped 563 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude Instant 1.2) vs 8k (Moonlight). Claude Instant 1.2 is closed, while Moonlight is open weight.

On HumanEval, Claude Instant 1.2 leads at 58.7% vs Moonlight at 48.1%. On GSM8K, Claude Instant 1.2 leads at 86.7% vs Moonlight at 77.4%.