Compare AI models

Specifications
Parameters
16B—
Context window
8k128k
API pricing
Input price
—$0.15
Output price
—$0.60
Cached input price
—$0.075
Cheapest input
—$0.15Azure
Cheapest output
—$0.60Azure
—$60.00OpenAI list price
Benchmarks
HumanEval
48.1%87.2%
MATH
45.3%70.2%
GPQA Diamond
35.2%40.2%
MMLU
70%82%
Overview
CompanyMoonshot AIOpenAI
Release dateFeb 22 2025Jul 18 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GPT-4o mini leads Moonlight on 4 of the 4 benchmarks they both report (HumanEval, MATH, GPQA Diamond, MMLU). Only GPT-4o mini has a verified first-party price for an available API: $0.15 per million input tokens and $0.60 per million output tokens. No currently available pay-as-you-go API rate is tracked for Moonlight. GPT-4o mini shipped 219 days before Moonlight, so benchmark comparisons should account for the intervening progress.

Context windows are 8k (Moonlight) vs 128k (GPT-4o mini). Moonlight is open weight, while GPT-4o mini is closed.

On HumanEval, GPT-4o mini leads at 87.2% vs Moonlight at 48.1%. On MATH, GPT-4o mini leads at 70.2% vs Moonlight at 45.3%. On GPQA Diamond, GPT-4o mini leads at 40.2% vs Moonlight at 35.2%. On MMLU, GPT-4o mini leads at 82% vs Moonlight at 70%.