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.
Together AI's Jev-style classifier. A LoRA fine-tune of Qwen3.5-4B that reads a state, a question and 2 to 24 options and returns one option letter. Served on Together's platform; recipe published.
Decides
choice, score, noul
choice, classify, route
Architecture
startlux-decision
tev
Fine-tuned from
—
qwen/qwen3.5-4b
License
cc-by-nc-4.0
Unspecified — weights licence being finalised
Availability
Open weights
Open weights + hosted API
Hosted by
—
Together AI
Input price
—
$0.042/MTok
Decision accuracy
88.3%
88.0%
Calibration error
—
—
Valid action rate
—
—
Median latency
26 ms
—
p95 latency
—
—
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 startlux-decision and tev1?
startlux-decision is from StartLux and tev1 from Together AI. startlux-decision has open weights you can download and run; tev1 has open weights and a hosted API. Both answer choice questions. Only startlux-decision answers score and noul. Only tev1 answers classify and route. tev1 is the smaller model, at 4.0B parameters to 4.7B. startlux-decision is licensed cc-by-nc-4.0; tev1, other.
Which is more accurate, startlux-decision or tev1?
They report on different suites — startlux-decision 88.3% on JevBench public set (231 items; 204 correct), maker's run, tev1 88.0% on Together development set (reused, not held out) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, startlux-decision or tev1?
startlux-decision: Free (open weights). tev1: $0.042 / $0 per 1M, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Can I run startlux-decision or tev1 locally?
Yes, both: systemone pull startlux/startlux-decision and systemone pull together-ai/tev1 download the weights.
JevBench public set (231 items; 204 correct), maker's run
