A 144M-parameter System One model on the multilingual mmBERT-small encoder. Takes a state, a question and 2 to 20 options and returns one decision with probabilities for choice, score and yes/no questions, on a CPU.
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, classify, route
choice, score, noul
Architecture
julia
startlux-decision
Fine-tuned from
jhu-clsp/mmbert-small
—
License
apache-2.0
cc-by-nc-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
73.2%
88.3%
Calibration error
—
—
Valid action rate
—
—
Median latency
—
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 julia-1 and startlux-decision?
julia-1 is from Supersonic Labs and startlux-decision from StartLux. Both have open weights you can download and run. Both answer choice, score and noul questions. Only julia-1 answers classify and route. julia-1 is the smaller model, at 144M parameters to 4.7B. julia-1 is licensed apache-2.0; startlux-decision, cc-by-nc-4.0.
Which is more accurate, julia-1 or startlux-decision?
They report on different suites — julia-1 73.2% on typed-decisions test set (400 cases, 2,000 questions), 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, julia-1 or startlux-decision?
julia-1: Free (open weights). startlux-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run julia-1 or startlux-decision locally?
Yes, both: systemone pull supersonic-labs/julia-1 and systemone pull startlux/startlux-decision download the weights.
Evaluation suite
typed-decisions test set (400 cases, 2,000 questions)
JevBench public set (231 items; 204 correct), maker's run
