Metask Lab's calibrated decision model in 16 languages. Qwen3.5-4B with a merged LoRA trained with the Nimble candidate-logit objective; one forward pass and a softmax over at most 26 answer-letter logits, with one temperature per question type.
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
metask-jev
open-jev
Fine-tuned from
qwen/qwen3.5-4b
microsoft/deberta-v3-large
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
80.1%
85.4%
Calibration error
—
0.022
Valid action rate
—
—
Median latency
62.8 ms
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 metask-jev and open-jev-deberta-v3-large?
metask-jev is from Metask Lab 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 metask-jev answers route. metask-jev reads up to 4K tokens of state, against 512 tokens for open-jev-deberta-v3-large. open-jev-deberta-v3-large is the smaller model, at 434M parameters to 4.5B.
Which is more accurate, metask-jev or open-jev-deberta-v3-large?
They report on different suites — metask-jev 80.1% on JevBench v1.2 public set (231 items) at 4,096 tokens, maker's run, 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, metask-jev or open-jev-deberta-v3-large?
metask-jev: Free (open weights). open-jev-deberta-v3-large: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Which is faster, metask-jev or open-jev-deberta-v3-large?
By their publishers’ figures, open-jev-deberta-v3-large answers in about 28 ms at the median and metask-jev in about 62.8 ms — measured on different hardware, so treat it as a rough guide.
Can I run metask-jev or open-jev-deberta-v3-large locally?
Yes, both: systemone pull metask-lab/metask-jev and systemone pull kotoba-labs/open-jev-deberta-v3-large download the weights.
Evaluation suite
JevBench v1.2 public set (231 items) at 4,096 tokens, maker's run
Kotoba held-out test, seen question types (banking77, SST-5, BoolQ)
