Compare AI models

Specifications
Parameters
671B30B
Context window
128k1M
API pricing
Cheapest input
—$0.00Nvidia
Cheapest output
—$0.00Nvidia
—$11.00Darkbloom · int4
Benchmarks
SWE-Bench Verified
49.2%51.6%
GPQA Diamond
71.5%75.4%
Overview
CompanyDeepSeekNVIDIA
Release dateJan 20 2025Aug 11 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Nemotron 3.5 Lightning leads DeepSeek R1 on 2 of the 2 benchmarks they both report (SWE-Bench Verified, GPQA Diamond). DeepSeek R1 shipped 568 days before Nemotron 3.5 Lightning, so benchmark comparisons should account for the intervening progress.

DeepSeek R1 has 671B parameters, while Nemotron 3.5 Lightning has 30B. Context windows are 128k (DeepSeek R1) vs 1M (Nemotron 3.5 Lightning).

On SWE-Bench Verified, Nemotron 3.5 Lightning leads at 51.6% vs DeepSeek R1 at 49.2%. On GPQA Diamond, Nemotron 3.5 Lightning leads at 75.4% vs DeepSeek R1 at 71.5%.