Nemotron Super
nvidia/nemotron-superFree, open reasoning engineered for maximum inference efficiency on NVIDIA hardware.
128K ctx · Free
NVIDIA
Open reasoning models tuned for its stack
Model
1
Founded
1993
Country
United States
Santa Clara
Founded 1993. Nemotron open-model programme from 2024.
NVIDIA builds its models out of Santa Clara, United States. Webparam tracks a single model from the lab: Nemotron Super, a reasoning model: free to call and a 128K . It ships with .
Models
The NVIDIA catalogue
One model, deliberately
NVIDIA lists one model on Webparam, and that is a strategy rather than a shortage: Nemotron Super is the reference implementation the whole programme points at. Nothing else on this page scales with catalogue size: the history, the releases and the research below are the same weight they would be for a lab shipping twenty.
Recent releases
What shipped lately
History
NVIDIA's Nemotron programme releases open reasoning models engineered to run superbly on its own hardware, free to use, meticulously optimised, and intended to grow the total market for accelerated inference. One model, deliberately: a reference implementation of efficient open reasoning.
NVIDIA's Nemotron programme releases open reasoning models engineered to run superbly on its own hardware, free to use, meticulously optimised, and intended to grow the total market for accelerated inference. One model, deliberately: a reference implementation of efficient open reasoning.
Strengths
- Free, open, hardware-optimised reasoning
- Reference-quality efficiency engineering
- Backed by the industry's compute leader
Use NVIDIA models through Webparam
Every model above is callable through one OpenAI-compatible convention: change the model ID, keep the request.
Related providers