NVIDIA releases Nemotron 3.5 Bolt 74B MoE that fits in 24GB VRAM at 2-bit
NerdyRodent · x · 2026-10-16
NVIDIA has published NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB on Hugging Face: a 74B-parameter MoE model with 9B active parameters, shipped as a 2-bit quantized build that fits entirely in 24GB of VRAM. It uses a standard chat template and supports multimodal (image/video/text) input. A compelling option for running a large MoE model locally or on a single GPU.
Related event: NVIDIA Open-Sources Nemotron 3.5 Super VL 120B Multimodal Model Family(33 posts)→
More from Infra
- Polymarket prices 28% odds a US state enacts data center moratorium by year-end — Polymarket · 2026-10-16
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- NVIDIA releases Qwen3.8-Flash-Next FP4 local model with new lossless compression format — TheZachMueller · 2026-10-16
- PyTorch now runs natively on AWS Trainium with zero code changes — PyTorch · 2026-10-16
- Chinese labs run on ~1/10 US compute; ByteDance pays 2-4x to hoover up SEA supply — robleclerc · 2026-10-16
- Zymtrace raises $12.2M and hires ML engineers to build agents that fix GPU bottlenecks — cfregly · 2026-10-16