← All model drops
Live drop
Nemotron 3.5 Bolt 74B
NVIDIADropped Oct 15, 2026
Compressed open MoE agent model optimized for local 24GB hardware with multimodal inputs.
About this release
Live drop
NVIDIADropped Oct 15, 2026
Compressed open MoE agent model optimized for local 24GB hardware with multimodal inputs.
About this release
Nemotron 3.5 Bolt 74B-A9B is a quantized open-weights edge variant of the Nemotron 3.5 Super VL architecture, released October 15, 2026 by NVIDIA. It activates 9B parameters and fits in 24GB VRAM while retaining text, image, and video understanding for local agentic workloads. Checkpoints and inference support are provided for llama.cpp, Ollama, vLLM and similar tools.
Read the official NVIDIA launch pageFirst builds incoming — this lane fills in automatically as people start building with Nemotron 3.5 Bolt 74B. Know one? Submit the URL.