Nvidia logo

Llama 3.1 Nemotron Ultra 253B v1

Nvidia

Overview

A 253B parameter derivative of Meta Llama 3.1 405B Instruct, developed by NVIDIA using Neural Architecture Search (NAS) and vertical compression. It underwent multi-phase post-training (SFT for Math, Code, Reasoning, Chat, Tool Calling; RL with GRPO) to enhance reasoning and instruction-following. Optimized for accuracy/efficiency tradeoff on NVIDIA GPUs. Supports 128k context.

Llama 3.1 Nemotron Ultra 253B v1 was released on April 7, 2025.

Performance

Timeline

ReleasedUnknown
Knowledge CutoffUnknown

Specifications

Parameters
253.0B
License
Llama 3.1 Community License
Training Data
Unknown

Benchmarks

Llama 3.1 Nemotron Ultra 253B v1 Performance Across Datasets

Scores sourced from the model's scorecard, paper, or official blog posts

LLM Stats Logollm-stats.com - Tue Jan 13 2026
Notice missing or incorrect data?Start an Issue discussion→

Pricing

Pricing, performance, and capabilities for Llama 3.1 Nemotron Ultra 253B v1 across different providers:

No pricing information available for this model.

Save up to 30% on AI inference costs

For companies with AI products in production. Limited spots available.

Apply Now

API Access

API Access Coming Soon

API access for Llama 3.1 Nemotron Ultra 253B v1 will be available soon through our gateway.

Recent Posts

Recent Reviews

FAQ

Common questions about Llama 3.1 Nemotron Ultra 253B v1

Llama 3.1 Nemotron Ultra 253B v1 was released on April 7, 2025.
Llama 3.1 Nemotron Ultra 253B v1 has 253.0 billion parameters.