Compare AI models

Specifications
Parameters
8B—
Context window
4k8k
API pricing
Cheapest input
—$30.00Azure
Cheapest output
—$60.00Azure
—$9,000Azure
Benchmarks
HellaSwag
76.4%95.3%
WinoGrande
70.9%87.5%
MMLU
54.4%86.4%
Overview
CompanyNVIDIAOpenAI
Release dateNov 15 2023Mar 14 2023
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

GPT-4 leads Nemotron-3 8B on 3 of the 3 benchmarks they both report (HellaSwag, WinoGrande, MMLU). GPT-4 shipped 246 days before Nemotron-3 8B, so benchmark comparisons should account for the intervening progress.

Context windows are 4k (Nemotron-3 8B) vs 8k (GPT-4). Nemotron-3 8B is open weight, while GPT-4 is closed.

On HellaSwag, GPT-4 leads at 95.3% vs Nemotron-3 8B at 76.4%. On WinoGrande, GPT-4 leads at 87.5% vs Nemotron-3 8B at 70.9%. On MMLU, GPT-4 leads at 86.4% vs Nemotron-3 8B at 54.4%.