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.
Respan's behaviour-scoring model for evals, guardrails and monitoring. For each plain-language behaviour you define, it reads a conversation or agent trace and returns the probability the behaviour is present, absent or not observable, in one forward pass.
Decides
choice, score, noul
noul, classify
Architecture
djev
span
Fine-tuned from
google/diffusiongemma-26b-a4b-it
—
License
apache-2.0
proprietary
Availability
Open weights + hosted API
Hosted API
Hosted by
Maisa
Respan
Input price
$0.035/MTok
$0.020/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 span-01?
djev is from Maisa and span-01 from Respan. djev has open weights and a hosted API; span-01 is only available as a hosted API. Both answer noul questions. Only djev answers choice and score. Only span-01 answers classify. djev is licensed apache-2.0; span-01, proprietary.
Which is more accurate, djev or span-01?
Neither publishes an accuracy figure. Test both on your own labelled examples.
Which is cheaper, djev or span-01?
djev: $0.035 / $0 per 1M, or free to self-host. span-01: $0.02 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run djev or span-01 locally?
djev yes — systemone pull maisa/djev downloads its weights. The other is only served as a hosted API.
