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.
Standard Thinking's decision model. Ministral 3 8B with a merged LoRA, served through stock SGLang and an open jev-adapter that exposes /v1/systemone and scores the supplied options in one forward pass with per-type temperatures.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
djev
standard-one
Fine-tuned from
google/diffusiongemma-26b-a4b-it
mistralai/ministral-3-8b-instruct-2512-bf16
License
apache-2.0
apache-2.0
Availability
Open weights + hosted API
Open weights + hosted API
Hosted by
Maisa
Standard Thinking
Input price
$0.035/MTok
—
Decision accuracy
—
71.1%
Calibration error
—
—
Valid action rate
—
—
Median latency
—
25.8 ms
p95 latency
—
41.9 ms
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 standard-one?
djev is from Maisa and standard-one from Standard Thinking. Both have open weights and a hosted API. Both answer choice, score and noul questions. Only standard-one answers classify and route. standard-one is the smaller model, at 8.0B parameters to 26B.
Which is more accurate, djev or standard-one?
Only standard-one publishes an accuracy figure (71.1% on typed-decisions suite, 400 cases / 2,000 decisions (maker's served run, v2)); djev does not, so there is no comparison to make without your own test.
Which is cheaper, djev or standard-one?
djev: $0.035 / $0 per 1M, or free to self-host. standard-one: Hosted, price not published, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Can I run djev or standard-one locally?
Yes, both: systemone pull maisa/djev and systemone pull standard-thinking/standard-one download the weights.
