NVIDIA
Nemotron
NVIDIA's research and instruction-tuning effort built on top of Llama base models, with strong RLHF and reward-modelling work.
Visit homepage ↗History & context
NVIDIA's Nemotron line builds on Llama base models with additional RLHF, reward modelling, and synthetic data generation. Llama 3.1 Nemotron 70B Instruct (October 2024) topped several Arena-style human-preference leaderboards on release.
Nemotron-4 340B (June 2024) was NVIDIA's larger synthetic-data-generation specialist — trained primarily to be a teacher for downstream instruction-tuning datasets rather than a chat-first model.
Inheriting Llama bases means inheriting the Llama Community License: free commercial use under 700M MAU.
Flagship model
2 models in this family
NVIDIA's reward-modelling research vehicle. Trained primarily to be a synthetic-data-generation specialist rather than a chat-first model. Useful for teams building instruction-tuning datasets at scale.
- Context
- 4K
- License
- llama-3
- VRAM Q4
- 204 GB
NVIDIA's RLHF-tuned Llama 3.1 70B. Tops several Arena-style human-preference leaderboards and shipped with NVIDIA's reward-model research. Inherits the Llama 3 community licence.
- Context
- 128K
- License
- llama-3
- VRAM Q4
- 42 GB