Compare AI models

Specifications
Parameters
16B—
Context window
8k200k
API pricing
Input price
—$2.00
Output price
—$8.00
Cached input price
—$0.50
Cheapest input
—$2.00OpenAI
Cheapest output
—$8.00OpenAI
—$800OpenAI list price
Benchmarks
HumanEval
48.1%87.4%
GPQA Diamond
35.2%82.8%
SimpleQA
5.6%49.4%
MMLU
70%92.9%
Overview
CompanyMoonshot AIOpenAI
Release dateFeb 22 2025Apr 16 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

o3 leads Moonlight on 4 of the 4 benchmarks they both report (HumanEval, GPQA Diamond, SimpleQA, MMLU). Only o3 has a verified first-party price for an available API: $2.00 per million input tokens and $8.00 per million output tokens. No currently available pay-as-you-go API rate is tracked for Moonlight. Moonlight shipped 53 days before o3, so benchmark comparisons should account for the intervening progress.

Context windows are 8k (Moonlight) vs 200k (o3). Moonlight is open weight, while o3 is closed.

On HumanEval, o3 leads at 87.4% vs Moonlight at 48.1%. On GPQA Diamond, o3 leads at 82.8% vs Moonlight at 35.2%. On SimpleQA, o3 leads at 49.4% vs Moonlight at 5.6%. On MMLU, o3 leads at 92.9% vs Moonlight at 70%.