Berget AI's System One model for gating agent commands: a LoRA adapter and a fine-tuned joint schema head on Cloudflare's Clef-Flash that answer noul, choice and score questions over a state in one forward pass. Trained on Swedish and English operations decisions.
StartLux's decision family in six sizes, 0.8B to a 35B-A3B mixture of experts. All questions in a request are answered in one forward pass, with a probability for every option, through a TypeSafe /v1/systemone-compatible server. Text, JSON or images, up to 256K tokens.
Decides
noul, choice, score
choice, score, noul
Architecture
clef
startlux-decision
Fine-tuned from
cloudflare/clef-flash
—
License
apache-2.0
cc-by-nc-4.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
97.0%
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 bev and startlux-decision?
bev is from Berget AI and startlux-decision from StartLux. Both have open weights you can download and run. Both answer noul, choice and score questions. startlux-decision is the smaller model, at 4.7B parameters to 9.0B. bev is licensed apache-2.0; startlux-decision, cc-by-nc-4.0.
Which is more accurate, bev or startlux-decision?
They report on different suites — bev 97.0% on Berget held-out risk split (16,902 questions; same operations-traffic corpora as training), 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, bev or startlux-decision?
bev: Free (open weights). startlux-decision: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run bev or startlux-decision locally?
Yes, both: systemone pull berget-ai/bev and systemone pull startlux/startlux-decision download the weights.
—
Evaluation suite
Berget held-out risk split (16,902 questions; same operations-traffic corpora as training)
JevBench public set (231 items; 204 correct), maker's run
