Native MLX FP16 conversion of Laya for Apple silicon. The same ModernBERT-large encoder, decision transformer and heads, running in MLX with no PyTorch; the checkpoint Laya Studio fine-tunes from.
A 0.6B replica of the System One idea trained on game environments. A Qwen3-0.6B backbone with an attention-based Choice head that scores a dynamic candidate set for Maze, Snake, ViZDoom and position prediction, shipped with its full training pipeline.
Decides
choice, score, noul
choice, noul, score
Architecture
laya
nanojev
Fine-tuned from
convai-innovations/laya
qwen/qwen3-0.6b
License
apache-2.0
mit
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
Questions
What is the difference between laya-mlx and nanojev?
laya-mlx is from aac6fef and nanojev from TianyuCodings. Both have open weights you can download and run. Both answer choice, score and noul questions. laya-mlx is the smaller model, at 421M parameters to 600M. laya-mlx is licensed apache-2.0; nanojev, mit.
Which is more accurate, laya-mlx or nanojev?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, laya-mlx or nanojev?
laya-mlx: Free (open weights). nanojev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run laya-mlx or nanojev locally?
Yes, both: systemone pull aac6fef/laya-mlx and systemone pull tianyu-codings/nanojev download the weights.
