by Nvidia
Llama-3.1-Nemotron-Ultra-253B is a 253 billion parameter reasoning-focused language model optimized for efficiency that excels at math, coding, and general instruction-following tasks while running on a single 8xH100 node.
| Input | $0.5 / 1M tokens |
| Output | $0.5 / 1M tokens |
Use nvidia/Llama-3_1-Nemotron-Ultra-253B-v1 via the AIHubMix unified API — one interface for every major LLM.