NVIDIA
NVIDIA develops the Nemotron model family alongside its computing hardware and software. The family includes models trained by NVIDIA and models adapted from other developers' releases. The entries identify the reviewed weights and their specific purpose.
Company & leadership
- President and CEO
- Jensen Huang
Founders
- Jensen Huang, Chris Malachowsky and Curtis Priem
From computer graphics to AI computing
Jensen Huang, Chris Malachowsky and Curtis Priem founded NVIDIA in 1993 for the 3D graphics market. Its corporate history marks the GPU launch in 1999 as a major step. Huang has served as president and CEO since the company began.
A hardware company that also publishes models
NVIDIA Corporation is based in Santa Clara. It incorporated in California in 1993 and reincorporated in Delaware in 1998. Its common stock trades on Nasdaq as NVDA. Nemotron is part of NVIDIA's model work; the presence of another company's model in NVIDIA's hosting catalog does not transfer authorship to NVIDIA.
What the Nemotron selection covers
Practical guidance. This page covers the reviewed Nemotron 3.5 releases and major Nemotron 3 reasoning and multimodal checkpoints. It does not enumerate every precision conversion, base checkpoint, reward model, experimental lab release or older Llama-Nemotron adaptation. Cosmos, robotics and unrelated models hosted in NIM have separate purposes and are outside this selection.
Separate the checkpoint from the service
Practical guidance. The entries describe published weights. Their availability does not promise a managed endpoint, free API capacity or a particular machine's performance. Check the chosen serving recipe, context length and precision together. Active expert parameters describe computation per token; all expert weights can still contribute to storage and memory needs.