A typed-decision layer for Google's DiffusionGemma, from David Villalón at Maisa AI. It compiles a request into a small answer canvas, runs one denoising read on patched vLLM and reads the probabilities of the allowed labels, for text, images and images offered as options.
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.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
djev
laya
Fine-tuned from
google/diffusiongemma-26b-a4b-it
answerdotai/modernbert-large
License
apache-2.0
apache-2.0
Availability
Open weights + hosted API
Open weights
Hosted by
Maisa
—
Input price
$0.035/MTok
—
Decision accuracy
—
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
39.5 ms
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 djev and laya?
djev is from Maisa and laya from Convai Innovations. djev has open weights and a hosted API; laya has open weights you can download and run. Both answer choice, score and noul questions. Only laya answers classify and route. laya is the smaller model, at 421M parameters to 26B.
Which is more accurate, djev or laya?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, djev or laya?
djev: $0.035 / $0 per 1M, or free to self-host. laya: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run djev or laya locally?
Yes, both: systemone pull maisa/djev and systemone pull convai-innovations/laya download the weights.
