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.
Respan's behaviour-scoring model for evals, guardrails and monitoring. For each plain-language behaviour you define, it reads a conversation or agent trace and returns the probability the behaviour is present, absent or not observable, in one forward pass.
Decides
choice, score, noul, classify
noul, classify
Architecture
open-jev
span
Fine-tuned from
microsoft/deberta-v3-large
—
License
apache-2.0
proprietary
Availability
Open weights
Hosted API
Hosted by
—
Respan
Input price
—
$0.020/MTok
Decision accuracy
85.4%
—
Calibration error
0.022
—
Valid action rate
—
—
Median latency
28 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 open-jev-deberta-v3-large and span-01?
open-jev-deberta-v3-large is from Kotoba Labs and span-01 from Respan. open-jev-deberta-v3-large has open weights you can download and run; span-01 is only available as a hosted API. Both answer noul and classify questions. Only open-jev-deberta-v3-large answers choice and score. open-jev-deberta-v3-large is licensed apache-2.0; span-01, proprietary.
Which is more accurate, open-jev-deberta-v3-large or span-01?
Only open-jev-deberta-v3-large publishes an accuracy figure (85.4% on Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)); span-01 does not, so there is no comparison to make without your own test.
Which is cheaper, open-jev-deberta-v3-large or span-01?
open-jev-deberta-v3-large: Free (open weights). span-01: $0.02 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run open-jev-deberta-v3-large or span-01 locally?
open-jev-deberta-v3-large yes — systemone pull kotoba-labs/open-jev-deberta-v3-large downloads its weights. The other is only served as a hosted API.
Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)
