System 1 ANE: millisecond AI decisions on Apple's Neural Engine with no text generation
pcuenq · x · 2026-10-10
The open-source system1-ane project (MIT) runs small task-specific decision models on the Apple Neural Engine via Core AI: a single forward pass returns a calibrated probability for each option — no generated text to parse, in a "System 1" style.
- Measured latency: 62-65ms for Snake/routing decisions, 0.3s for Tetris on an M5 Max.
- Runs a Qwen3.5-0.8B fine-tune (Jeff) plus Unsloth-trained adapters.
- Ships Tetris, Snake and text-routing demos with training notebooks.
More from Infra
- Micron didn't end up with zero Rubin HBM, as analyst commentary backpedals — suchenzang · 2026-10-10
- oMLX 0.7.1.dev1 adds decision models and batched prefill, +31% Qwen3.6 decode speed on M5 Max — pcuenq · 2026-10-10
- Bittensor GPU rental network sees 45% spend growth, 106% more rentals in monthly report — markjeffrey · 2026-10-10
- Microsoft doubles down on local AI with Nvidia RTX Spark Surface Ultra priced up to $5,899 — MooseEfficient2151 · 2026-10-10
- Building a pit crew for Grok Bot: frontier model plans, free models grind — alexcovo_eth · 2026-10-10
- AI boom turns into a debt boom: Oracle 5y CDS near record 261bps, implying 20.4% default odds — cyb3rops · 2026-10-10