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.
StartLux's decision family in five sizes, 0.8B to 27B. All questions in a request are answered in one forward pass, with a probability for every option, through a TypeSafe /v1/systemone-compatible server that records CUDA graphs for short requests.
Decides
choice, score, noul
choice, score, noul
Architecture
cygnet
startlux-decision
Fine-tuned from
google/gemma-4-12b-it
—
License
mit
cc-by-nc-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
87.9%
88.3%
Calibration error
—
—
Valid action rate
—
—
Median latency
50 ms
26 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 startlux-decision?
cygnet is from Blockbrain Labs and startlux-decision from StartLux. Both have open weights you can download and run. Both answer choice, score and noul questions. startlux-decision is the smaller model, at 4.7B parameters to 12B. cygnet is licensed mit; startlux-decision, cc-by-nc-4.0.
Which is more accurate, cygnet or startlux-decision?
They report on different suites — cygnet 87.9% on JevBench public set (231 items), JevBench CLI, startlux-decision 88.3% on JevBench public set (231 items; 204 correct), maker's run — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, cygnet or startlux-decision?
cygnet: Free (open weights). startlux-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Which is faster, cygnet or startlux-decision?
By their publishers’ figures, startlux-decision answers in about 26 ms at the median and cygnet in about 50 ms — measured on different hardware, so treat it as a rough guide.
Can I run cygnet or startlux-decision locally?
Yes, both: systemone pull blockbrain-labs/cygnet and systemone pull startlux/startlux-decision download the weights.
Evaluation suite
JevBench public set (231 items), JevBench CLI
JevBench public set (231 items; 204 correct), maker's run
