Qwen3.5-4B with a merged rank-16 LoRA distilled from 17,408 teacher questions and 30,000 public training rows, read out as a temperature-scaled softmax over answer-letter logits. Also in 9B, 2B and GGUF.
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, classify, route
choice, classify, route
Architecture
jevk5
tev
Fine-tuned from
qwen/qwen3.5-4b
qwen/qwen3.5-4b
License
apache-2.0
Unspecified — weights licence being finalised
Availability
Open weights
Open weights + hosted API
Hosted by
—
Together AI
Input price
—
$0.042/MTok
Decision accuracy
78.4%
88.0%
Calibration error
0.035
—
Valid action rate
—
—
Median latency
13.2 ms
—
p95 latency
—
—
Evaluation suite
JevBench public hard tier
Together development set (reused, not held out)
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 jevk5 and tev1?
jevk5 is from Alibi Serikbay and tev1 from Together AI. jevk5 has open weights you can download and run; tev1 has open weights and a hosted API. Both answer choice, classify and route questions. Only jevk5 answers score and noul. jevk5 is licensed apache-2.0; tev1, other.
Which is more accurate, jevk5 or tev1?
They report on different suites — jevk5 78.4% on JevBench public hard tier, 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, jevk5 or tev1?
jevk5: 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 jevk5 or tev1 locally?
Yes, both: systemone pull alibi-serikbay/jevk5 and systemone pull together-ai/tev1 download the weights.
