NVIDIA: Llama 3.3 Nemotron Super 49B V1.5
nvidia/llama-3.3-nemotron-super-49b-v1.5
Reasoning
Overview
Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context. It’s post-trained for agentic workflows (RAG, tool calling) via SFT across math, code, science, and...
Code example
Related models
Other models from the same provider