Compare AI models

Specifications
Parameters
51B21B
Context window
128k128k
API pricing
Cheapest input
—$0.018AkashML
Cheapest output
—$0.09AkashML
—$8.10Darkbloom · fp8
Benchmarks
MMLU
80.2%85.3%
Overview
CompanyNVIDIAOpenAI
Release dateSep 23 2024Aug 5 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

gpt-oss-20b leads Llama 3.1 Nemotron 51B on 1 of the 1 benchmark they both report (MMLU). Llama 3.1 Nemotron 51B shipped 316 days before gpt-oss-20b, so benchmark comparisons should account for the intervening progress.

Llama 3.1 Nemotron 51B has 51B parameters, while gpt-oss-20b has 21B. Context windows are 128k (Llama 3.1 Nemotron 51B) vs 128k (gpt-oss-20b).

On MMLU, gpt-oss-20b leads at 85.3% vs Llama 3.1 Nemotron 51B at 80.2%.