Ngomhla ka-11 Agasti ngesikhathi sendawo, i-NVIDIA imemezele ukwethulwa kwemodeli ye-AI yomthombo ovulekile engu-30B, i-Nemotron 3.5 Lightning, yandisa uchungechunge lwayo lwamamodeli e-Nemotron 3.
Ngokwesitatimende esisemthethweni, i-Nemotron 3.5 Lightning iyimodeli ye-MoE enamapharamitha ayizigidi eziyizinkulungwane ezingu-30, eyakhelwe ngokukhethekile imisebenzi ethile ezinhlelweni ezinkulu ezinama-agent amaningi, okusiza ekuthuthukiseni izinhlelo zama-agent ezihlakaniphe nezisebenza kahle.
Uma iqhathaniswa namanye amamodeli afanayo, le modeli ingakhuphula isivinini sokukhiqiza ngokuphindwe ka-4, ngaleyo ndlela isheshise isivinini sonke sokuqedwa kwemisebenzi yama-agent ngo-30%.
I-Nemotron 3.5 Lightning ingasebenza ngokuqondile ezinhlelweni ze-AI zasendaweni, okuhlanganisa i-NVIDIA RTX PC, i-NVIDIA DGX Spark, i-NVIDIA DGX Station kanye ne-NVIDIA Jetson; ingasetshenziswa nasezimweni zezinhlelo zebhizinisi, futhi inwetshwe ngaphandle komthungo iye kumadivayisi e-edge AI, eziteshini zokusebenza ze-NVIDIA RTX PRO, ezikhungweni zedatha nasezindaweni zamafu.
Izixhumanisi zomthombo ovulekile:
Hugging Face: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4
ModelScope: https://modelscope.ai/collections/nv-community/Nemotron-35-Lightning
OpenRouter: https://openrouter.ai/nvidia/nemotron-3.5-lightning:free
NVIDIA: https://build.nvidia.com/nvidia/nemotron-3.5-lightning-30b-a3b