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.
Juspay's open decision model. Qwen3.6-35B-A3B with a merged rank-16 LoRA, shipped with a pinned SGLang serving bundle that answers Choice, Score and Noul questions over text, images or a video through a TypeSafe-compatible /v1/systemone API.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
djev
xor
Fine-tuned from
google/diffusiongemma-26b-a4b-it
qwen/qwen3.6-35b-a3b
License
apache-2.0
apache-2.0
Availability
Open weights + hosted API
Open weights
Hosted by
Maisa
—
Input price
$0.035/MTok
—
Decision accuracy
—
90.0%
Calibration error
—
0.073
Valid action rate
—
—
Median latency
—
69 ms
p95 latency
—
162 ms
Evaluation suite
—
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 xor?
djev is from Maisa and xor from Juspay. djev has open weights and a hosted API; xor has open weights you can download and run. Both answer choice, score and noul questions. Only xor answers classify and route. djev is the smaller model, at 26B parameters to 35B.
Which is more accurate, djev or xor?
Only xor publishes an accuracy figure (90.0% on JevBench public set (231 items), maker's self-run of Xor 1.2); djev does not, so there is no comparison to make without your own test.
Which is cheaper, djev or xor?
djev: $0.035 / $0 per 1M, or free to self-host. xor: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run djev or xor locally?
Yes, both: systemone pull maisa/djev and systemone pull juspay/xor download the weights.
JevBench public set (231 items), maker's self-run of Xor 1.2
