Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models

blogs.nvidia.comModels4dModels4d

NVIDIA released three new open models in the Nemotron 3.5 family: Nemotron 3.5 Super VL, a 120-billion-parameter model for enterprise servers, and two edge models, Nemotron 3.5 Bolt 74B-A9B 24GB and Nemotron 3.5 Bolt E4B 4GB. Super VL and Bolt 74B-A9B are available now, with Bolt E4B coming Oct. 23. The models cover agent deployments from large enterprise servers to edge devices and local hardware.

Key sources

SOURCEOct 15, 10:00 PMdeveloper.nvidia.com
SOURCE@nvidiaaiOct 15, 10:09 PM“120B total parameters with 12B active”x.com
SUPPORT@togethercomputeOct 15, 10:01 PM“always-on agents that see, reason, and act across documents, screens, and video”x.com
SUPPORT@lambdaapiOct 15, 10:02 PM“One NVIDIA Blackwell GPU at NVFP4 out-throughputs two NVIDIA Blackwell GPUs at BF16 on the same 8,192-in / 1,024-out workload”x.com
SUPPORT@gordic_aleksaOct 15, 10:24 PM“Super scores 51.78% on SWE-bench Pro (better than Nemotron 3 Ultra despite ~4.6× fewer active params!)”x.com
SUPPORT@miaai_labOct 15, 9:57 PM“Weights ~75 GB, fits a solo DGX Spark”x.com
SUPPORT@unslothaiOct 15, 10:23 PM“NVIDIA releases 2 new models: Nemotron-3.5-Bolt and Nemotron-3.5-Super-VL”x.com
SOURCEhuggingnewshuggingnews.com

Markdown