NVIDIA Launches Nemotron 3.5 Bolt: 74B Model in 24GB VRAM
NVIDIA released Nemotron 3.5 Bolt on Hugging Face, a 74B-parameter (9B active) MoE vision-language model compressed to run in 24GB VRAM, enabling agent workloads on consumer RTX GPUs.
2026-10-16 ~ 2026-10-16 · 2 related posts
- Episode 1: NVIDIA Releases Nemotron 3.5: Super 120B and Lightning 30B(2026-10-15, 6 posts)
- Episode 2: NVIDIA Open-Sources Nemotron 3.5 Super VL 120B Multimodal Model Family(2026-10-15, 33 posts)
- Episode 3: NVIDIA Nemotron 3.5 Super VL tops US open-weight models(2026-10-15, 2 posts)
- Episode 4: NVIDIA Launches Nemotron 3.5 Bolt: 74B Model in 24GB VRAM(2026-10-16, 2 posts)
- NVIDIA ships Nemotron 3.5 Bolt: 74B-A9B model in 24GB of memory — NVIDIAAI · 2026-10-16
- NVIDIA ships Nemotron 3.5 Bolt: 74B MoE agent model compressed into 24 GB — lmoroney · 2026-10-16