Laya ported to MLX: M3 Max runs local agent 60 decisions/sec, 50x faster
sven_ai · x · 2026-09-21
A developer ported the Laya agent to Apple's MLX framework and ran it locally on an M3 Max, playing Snake with under 1GB of memory and 60 decisions per second — about 50x faster than Jev. A striking demonstration of how far Apple Silicon plus MLX can push lightweight local agents.
More from coding & agent
- User automates entire video clipping workflow with Nous Research's Hermes agent — Teknium · 2026-11-12
- As models improve, writing good evals gets harder and more important — amankhan · 2026-11-12
- Prime agent ships with day-one multi-agent communication, runs in Minecraft unmodified — sarahcat21 · 2026-11-12
- Autonomous agent edits just 1,024 weights to lift Qwen3.5 0.8B from 7.4% to 37.4% on MATH-500 — VictorSanh · 2026-11-12
- Omarchy 4.6 lets you share local AI models across your tailnet, says DHH — alexcovo_eth · 2026-11-12
- Coding harnesses hand shell access to inference providers — 4-step gateway mitigation — shensi · 2026-11-12