NVIDIA · ga · free+api
NVIDIA Nemotron
NVIDIA’s own models, tuned to run well on their chips and cloud NIM.
- Best for
- NIM microservices on NVIDIA GPUs
- Context
- 128K
- Modalities
- text
- Released
- 2024-10
Pro take
Use with NIM / NGC. Compare Llama Nemotron variants for reasoning vs chat. CUDA lock-in is the point.