Llama Nemotron Ultra 253B
Major release: A new general-purpose AI model from a leading lab, the kind behind ChatGPT, Claude or Gemini: a new flagship or generation.Open model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.
Key facts
- Released
- Mar 18, 2025
- What it does
- Text and chat: Writes, answers questions and holds a conversation.Code: Writes, explains and fixes code.Thinks first: Can reason step by step before it answers: slower, but better at maths, code and tricky questions.
View moreView less
- How
- Free to download, with some licence limits
- Size
- Parameters: The numbers a model learns during training: a rough measure of its size. Bigger usually means more capable, but slower and pricier to run. 7B means 7 billion.
- Announced as
- “Ultra is 253B distilled from Llama 3.1 405B for maximum agentic accuracy on multi-GPU data center servers.”
From trckr's archive of releases before it started watching. Date and details from Epoch AI's database of AI models (CC BY 4.0).
Compare Llama Nemotron Ultra 253B with
It came out 57 days after Eagle 2, the lab's previous model.
The details
Get alerts like this the moment they happen.
From NVIDIA and every other major lab.
Free, no account needed. Push, Telegram, Discord, email or RSS.