Llama Nemotron Nano
Announced on nvidia.com
Benchmark results
| Benchmark | Score | Source check |
|---|---|---|
| AIME 2025No tools · reasoning on; average of up to 16 runs; 32K sequence length | 47.1% | |
| reasoning off; greedy decoding; 32K sequence length | 0% |
Sources
Reported by NVIDIA
- Llama-3.1-Nemotron-Nano-8B-v1 model card (Hugging Face, launch-day revision)huggingface.co · 18 Mar 2025