Nemotron 3 Ultra
NVIDIA released a very large free-to-download AI for reasoning and coordinating tasks, meant for companies and developers with powerful servers.
This came out before trckr began following OpenRouter, so it was added on Nov 6 without an alert. Its release date comes from OpenRouter.
Key facts
- Released
- Jun 4, 2026
- What it does
- Text and chat: Writes, answers questions and holds a conversation.Thinks first: Can reason step by step before it answers: slower, but better at maths, code and tricky questions.
Can I use it?
In an app
No app mentioned so far.
For developers
Developers can use it now through an API: How developers plug a model into their own apps and pay per use. A model that's "in the API" can be used by programmers, even if it isn't in a chat app yet..
Download
Free to download and run on your own computer: it has Open weights: The model's files are public, so anyone can download it, run it on their own computer and build on it, usually for free. trckr labels these "Open"..
Price and limits
What developers pay to use it through the API: How developers plug a model into their own apps and pay per use. A model that's "in the API" can be used by programmers, even if it isn't in a chat app yet., per million Tokens: The small chunks of text a model reads and writes, about three-quarters of a word each. Prices and limits are counted in tokens. sent (input) or written back (output), and how much it can take in at once.
- Input
- $0.60
- per million tokensCached: $0.12
- Output
- $2.40
- per million tokens
- Context window: How much a model can keep in mind at once: your question, any files you share and its own replies. Measured in tokens; 1,000 tokens is about 750 words.
- 203K tokens
- about 150,000 words
From OpenRouter, checked .
How it scores
From independent testers, never the lab's own claims.
- Combined test score#79 of 235
- GPT-6 Astra169
- 2GPT-6.1 Sol169
- 78Qwen 3.5 Plus147
- 79Nemotron 3 Ultra147
- 80o3147
Source: Epoch AI · Epoch Capabilities Index
Compare Nemotron 3 Ultra with
It came out 85 days after Nemotron 3 Super, the lab's previous model.
The details
NVIDIA's open frontier-reasoning and orchestration model, with 55B active out of 550B total in a hybrid Transformer-Mamba Mixture of experts (MoE): A design that splits a big model into many "expert" parts and uses only a few for each word. It gives big-model quality at closer to small-model speed. design. It is a very large model that needs powerful servers to run.
Model ID: The exact name developers type to use a model, like gpt-5-mini. New IDs often show up in a lab's systems before the official announcement, which is how trckr spots many releases early.
nvidia/nemotron-3-ultra-550b-a55bTimeline
/OpenRouterFirst signal
Get alerts like this the moment they happen.
From NVIDIA and every other major lab.