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.
The first System One model. Reads a state, answers typed Choice, Score and Noul questions in one call with calibrated probabilities, and generates no text. Closed weights, served by TypeSafe AI.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
djev
jev
Fine-tuned from
google/diffusiongemma-26b-a4b-it
—
License
apache-2.0
proprietary
Availability
Open weights + hosted API
Hosted API
Hosted by
Maisa
TypeSafe AI
Input price
$0.035/MTok
$0.042/MTok
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 djev and jev?
djev is from Maisa and jev from TypeSafe AI. djev has open weights and a hosted API; jev is only available as a hosted API. Both answer choice, score and noul questions. Only jev answers classify and route. djev is licensed apache-2.0; jev, proprietary.
Which is more accurate, djev or jev?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, djev or jev?
djev: $0.035 / $0 per 1M, or free to self-host. jev: $0.042 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run djev or jev locally?
djev yes — systemone pull maisa/djev downloads its weights. The other is only served as a hosted API.
