Typed decisions from Google's frozen Gemma-4-12B-it, with no fine-tuning. A small shim over unmodified vLLM letters the options, reads the model's own probability for each letter at one answer position and applies one calibration temperature; one output token per decision.
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.
Decides
choice, score, noul
choice, score, noul
Architecture
cygnet
djev
Fine-tuned from
google/gemma-4-12b-it
google/diffusiongemma-26b-a4b-it
License
mit
apache-2.0
Availability
Open weights
Open weights + hosted API
Hosted by
—
Maisa
Input price
—
$0.035/MTok
Decision accuracy
87.9%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
50 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 cygnet and djev?
cygnet is from Blockbrain Labs and djev from Maisa. cygnet has open weights you can download and run; djev has open weights and a hosted API. Both answer choice, score and noul questions. cygnet is the smaller model, at 12B parameters to 26B. cygnet is licensed mit; djev, apache-2.0.
Which is more accurate, cygnet or djev?
Only cygnet publishes an accuracy figure (87.9% on JevBench public set (231 items), JevBench CLI); djev does not, so there is no comparison to make without your own test.
Which is cheaper, cygnet or djev?
cygnet: Free (open weights). djev: $0.035 / $0 per 1M, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Can I run cygnet or djev locally?
Yes, both: systemone pull blockbrain-labs/cygnet and systemone pull maisa/djev download the weights.
