NVIDIA's 74B Nemotron 3.5 Bolt, top US open-weight model, now runs locally on Mac
SumitGup · x · 2026-10-16
Rapid-MLX has quantized NVIDIA's new Nemotron 3.5 Bolt 74B-A9B (hybrid Mamba/MoE, 256K context) for Apple Silicon and published it on Hugging Face:
- 3-bit mixed (31.7 GB): requires 48GB+ unified memory
- 4-bit (43.3 GB): requires 64GB+ unified memory
Per cited Artificial Analysis data, the companion Nemotron 3.5 Super VL scores 26.5, making it the leading US open-weight model, ahead of Thinking Machines' Inkling Small (25.7) and NVIDIA's Nemotron 3 Ultra (23); Bolt 74B scores 21.
More from Infra
- Samsung cuts Q4 phone output as Nittobo shifts E-glass to AI-chip T-glass — zephyr_z9 · 2026-10-16
- White House AI chief discusses Empire AI, New York's shared compute infrastructure — mkratsios47 · 2026-10-16
- llama.cpp merges Muse Spark Open (816B-A42B) text, vision and audio support in 4 hours — jacek2023 · 2026-10-16
- FT warns of 'Darwin moment' for AI borrowers as CoreWeave bonds sink to distress levels — firstadopter · 2026-10-16
- Phone as interface, home servers as the lab: a two-workflow local AI setup — EAccelerate_42 · 2026-10-16
- "A datacenter is a bank": CoreWeave drama revives calls for算力 supervision — arthurcolle · 2026-10-16