AutoTrust's student of TypeSafe Jev 1.13. A frozen Qwen3.8-27B plus a 108.9M-parameter decision block trained on Jev's own output distributions; one set of weights answers typed questions in one pass (System 1) or generates text with the untouched base (System 2).
Small open typed-decision model with calibrated probabilities, first-class Nepali coverage.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
blocks-of-experts
mira
Fine-tuned from
qwen/qwen3.8-27b
jhu-clsp/mmbert-small
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
88.7%
74.5%
Calibration error
—
0.025
Valid action rate
—
100.0%
Median latency
137 ms
30 ms
p95 latency
—
43 ms
Evaluation suite
JevBench public set (231 items), family-macro score, maker's run
s1-decision-bench
Latest version
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 jev-27b and mira?
jev-27b is from AutoTrust AI Lab and mira from SAGEA. Both have open weights you can download and run. Both answer choice, score and noul questions. Only mira answers classify and route.
Which is more accurate, jev-27b or mira?
They report on different suites — jev-27b 88.7% on JevBench public set (231 items), family-macro score, maker's run, mira 74.5% on s1-decision-bench — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, jev-27b or mira?
jev-27b: Free (open weights). mira: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Which is faster, jev-27b or mira?
By their publishers’ figures, mira answers in about 30 ms at the median and jev-27b in about 137 ms — measured on different hardware, so treat it as a rough guide.
Can I run jev-27b or mira locally?
Yes, both: systemone pull autotrust-ai/jev-27b and systemone pull sagea/mira download the weights.
