Compare AI models

Specifications
Parameters
49B21B
Context window
128k128k
API pricing
Cheapest input
—$0.018AkashML
Cheapest output
—$0.09AkashML
—$8.10Darkbloom · fp8
Benchmarks
AIME 2024
87.5%92.1%
AIME 2025
82.71%91.7%
GPQA Diamond
71.97%71.5%
Overview
CompanyNVIDIAOpenAI
Release dateJul 25 2025Aug 5 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

gpt-oss-20b leads Llama Nemotron Super 49B v1.5 on 2 of the 3 benchmarks they both report (AIME 2024, AIME 2025, GPQA Diamond). Llama Nemotron Super 49B v1.5 shipped 11 days before gpt-oss-20b, so benchmark comparisons should account for the intervening progress.

Llama Nemotron Super 49B v1.5 has 49B parameters, while gpt-oss-20b has 21B. Context windows are 128k (Llama Nemotron Super 49B v1.5) vs 128k (gpt-oss-20b).

On AIME 2024, gpt-oss-20b leads at 92.1% vs Llama Nemotron Super 49B v1.5 at 87.5%. On AIME 2025, gpt-oss-20b leads at 91.7% vs Llama Nemotron Super 49B v1.5 at 82.71%. On GPQA Diamond, Llama Nemotron Super 49B v1.5 leads at 71.97% vs gpt-oss-20b at 71.5%.