LiquidAI d1-2B decision model runs fully local on M5 Max: ~14 decisions/s, zero tokens
iamrobotbear · x · 2026-11-17
Developer Ivan Fioravanti demoed LiquidAI's d1-2B decision model via mlx-vlm on an Apple Silicon M5 Max MacBook Pro: it reads gestures and expressions from a webcam in 68 ms, makes 14 decisions per second, generates zero tokens, and runs 100% locally — a showcase of small decision models for real-time on-device perception.
Related event: LiquidAI d1-2B Runs Locally: 14 Decisions Per Second With Zero Tokens(2 posts)→
More from Infra
- Altman: No chip shortage—supply chain wrongly assumes AI labs will buy everything — RihardJarc · 2026-11-17
- Azure Linux 4.0 debuts on a Fedora base as Microsoft says it's ready for mainstream workloads — unixterminal · 2026-11-17
- US shuts Malaysia chip loophole: license now needed to reexport advanced chips to five countries — SumitGup · 2026-11-17
- Gary Marcus mocks NVDA credit initiatives as 'the bank is now closed' — GaryMarcus · 2026-11-17
- Bharat Semi raises $24M seed to build programmable smart sensors for Physical AI — adityaag · 2026-11-17
- Hugging Face Hub billing now rides on Google Cloud invoices — rseroter · 2026-11-17