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.
An independent Jev-shaped reproduction on DeBERTa-v3-large. One state and any number of choice, score and yes/no questions go in, and a calibrated distribution per question comes out of one forward pass. Public gold labels only.
Decides
choice, score, noul, classify, route
choice, score, noul, classify
Architecture
julia
open-jev
Fine-tuned from
jhu-clsp/mmbert-small
microsoft/deberta-v3-large
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
73.2%
85.4%
Calibration error
—
0.022
Valid action rate
—
—
Median latency
—
28 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 open-jev-deberta-v3-large?
julia-1 is from Supersonic Labs and open-jev-deberta-v3-large from Kotoba Labs. Both have open weights you can download and run. Both answer choice, score, noul and classify questions. Only julia-1 answers route. julia-1 reads up to 8K tokens of state, against 512 tokens for open-jev-deberta-v3-large. julia-1 is the smaller model, at 144M parameters to 434M.
Which is more accurate, julia-1 or open-jev-deberta-v3-large?
They report on different suites — julia-1 73.2% on typed-decisions test set (400 cases, 2,000 questions), open-jev-deberta-v3-large 85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, julia-1 or open-jev-deberta-v3-large?
julia-1: Free (open weights). open-jev-deberta-v3-large: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run julia-1 or open-jev-deberta-v3-large locally?
Yes, both: systemone pull supersonic-labs/julia-1 and systemone pull kotoba-labs/open-jev-deberta-v3-large download the weights.
Evaluation suite
typed-decisions test set (400 cases, 2,000 questions)
Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)
