An open-weights System One model from Convai Innovations. A fully fine-tuned ModernBERT-large encoder with a from-scratch decision head that scores one marker per option and answers every question in a single 33–39 ms pass. Runs on your own CPU or GPU.
Unofficial NoulXP package of Convai Innovations' Laya (English) (Apache-2.0), fp32-o23: checked against the model's own answers, none changed. Not made by Convai Innovations.
Decides
choice, score, noul, classify, route
choice, score, noul, classify, route
Architecture
laya
laya
Fine-tuned from
answerdotai/modernbert-large
convai-innovations/laya
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
—
—
Calibration error
—
—
Valid action rate
—
—
Median latency
39.5 ms
116 ms
p95 latency
—
859 ms
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 and laya-noulxp-cpu?
laya is from Convai Innovations and laya-noulxp-cpu from Convai Innovations. Both have open weights you can download and run. Both answer choice, score, noul, classify and route questions. laya is the smaller model, at 421M parameters to 421M. laya-noulxp-cpu is fine-tuned from laya.
Which is more accurate, laya or laya-noulxp-cpu?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, laya or laya-noulxp-cpu?
laya: Free (open weights). laya-noulxp-cpu: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Which is faster, laya or laya-noulxp-cpu?
By their publishers’ figures, laya answers in about 39.5 ms at the median and laya-noulxp-cpu in about 116 ms — measured on different hardware, so treat it as a rough guide.
Can I run laya or laya-noulxp-cpu locally?
Yes, both: systemone pull convai-innovations/laya and systemone pull convai-innovations/laya-noulxp-cpu download the weights.
Evaluation suite
—
Measured by System One Models on 2026-10-05: one question per request, 4 threads of a Xeon Gold 6342
