Compare AI models

Specifications
Parameters
671B120B
Context window
128k1M
API pricing
Cheapest input
$0.256DeepInfra$0.00Nvidia
Cheapest output
$0.712DeepInfra$0.00Nvidia
$103StreamLake$33.00Io Net · nvfp4
Benchmarks
SWE-Bench Verified
42%71.1%
LiveCodeBench
40.5%88.22%
GPQA Diamond
59.1%89.52%
Overview
CompanyDeepSeekNVIDIA
Release dateDec 26 2024Oct 15 2026
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Nemotron 3.5 Super VL leads DeepSeek V3 on 3 of the 3 benchmarks they both report (SWE-Bench Verified, LiveCodeBench, GPQA Diamond). DeepSeek V3 shipped 658 days before Nemotron 3.5 Super VL, so benchmark comparisons should account for the intervening progress.

DeepSeek V3 has 671B parameters, while Nemotron 3.5 Super VL has 120B. Context windows are 128k (DeepSeek V3) vs 1M (Nemotron 3.5 Super VL).

On SWE-Bench Verified, Nemotron 3.5 Super VL leads at 71.1% vs DeepSeek V3 at 42%. On LiveCodeBench, Nemotron 3.5 Super VL leads at 88.22% vs DeepSeek V3 at 40.5%. On GPQA Diamond, Nemotron 3.5 Super VL leads at 89.52% vs DeepSeek V3 at 59.1%.