d1-ane
by nathanrchn
Converts and benchmarks decision models on Apple Neural Engine, with a Tetris demo
LiquidAI's d1-omni-600M decision model on the Apple Neural Engine, with a Tetris demo
On X
d1-omni-600M plays Tetris against Laya at 4 ms per decision on M4 Max
Same Tetris harness with @liquidai's d1-omni-600M on ANE, ~4 ms per decision on an M4 Max. Same speed. Same pieces for both. Laya (@Nandakishorm1) is done after 350. d1 places 22,832 and clears 9,118 lines before it tops out. Try out yourself:
Project: d1-ane→Related projects
colibri
JustVugg
Runs large open models locally in pure C, including Laya for typed decisions
40.1k·GitHub repo
laya
NandhaKishorM
Runs multilingual typed-decision models and provides routing, serving, and integration options
23.9k·GitHub repo
laya-mlx
mizorewww
Runs Laya typed-decision models locally on Apple Silicon using MLX
6.3k·GitHub repo
C
laya
convaiinnovations
A multilingual-ready System 1 model that returns calibrated typed decisions over text in one pass
45.1k·HF model
laya-coreml
mizorewww
Runs Laya typed-decision models locally on Apple Silicon using Core ML and the Neural Engine
1.4k·GitHub repo
laya
receptron
Runs Laya typed-decision models from Node.js and TypeScript through ONNX Runtime
450·GitHub repo