Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models
NVIDIA · 16 Oct 2026, 01:00 AEST
NVIDIA has expanded its Nemotron family with the release of the Nemotron 3.5 Super VL and two new edge models, Bolt 74B-A9B and Bolt E4B. These models are designed to optimize performance for various agent deployments, from enterprise servers to edge devices.
Key facts
- NVIDIA today released Nemotron 3.5 Super VL, an open 120-billion-parameter model that delivers the fastest task completion and highest accuracy for specialized always-on agents.
- Joining it are two new edge models: Nemotron 3.5 Bolt 74B-A9B 24GB, optimized to balance intelligence and speed per GB for local agentic workloads, and Nemotron 3.5 Bolt E4B 4GB, optimized for intelligence per GB for fast conversational agents.
- Super VL, built in collaboration with the Nemotron Coalition, delivers leading accuracy on multimodal agentic tasks — instruction following, tool use, multistep business workflows, and visual understanding — comparable with NVIDIA’s largest reasoning model at roughly one-fifth the parameter count.
- With Bolt 74B-A9B 24GB, users can delegate their projects to popular local agent harnesses — from researching the web to writing to debugging code — all without sending those workloads to the cloud.
Read aloud
NVIDIA has expanded its Nemotron family with the release of the Nemotron three point five Super VL and two new edge models, Bolt 74B-A9B and Bolt E4B. These models are designed to optimize performance for various agent deployments, from enterprise servers to edge devices. NVIDIA today released Nemotron three point five Super VL, an open one hundred twenty-billion-parameter model that delivers the fastest task completion and highest accuracy for specialized always-on agents. Joining it are two new edge models: Nemotron three point five Bolt 74B-A9B 24GB, optimized to balance intelligence and speed per GB for local agentic workloads, and Nemotron three point five Bolt E4B 4GB, optimized for intelligence per GB for fast conversational agents. Super VL, built in collaboration with the Nemotron Coalition, delivers leading accuracy on multimodal agentic tasks — instruction following, tool use, multistep business workflows, and visual understanding — comparable with NVIDIA’s largest reasoning model at roughly one-fifth the parameter count. With Bolt 74B-A9B 24GB, users can delegate their projects to popular local agent harnesses — from researching the web to writing to debugging code — all without sending those workloads to the cloud.
Sources in this cluster
- Open From Data Center to Desktop: NVIDIA Expands Nemotron Family With Highest Efficiency Super VL and Bolt Models
NVIDIA · 16 Oct 2026, 01:00 AEST