An embedding model turned decision model. Qwen3-Embedding-4B with a merged LoRA, driven by the open JevEmbed toolkit, which scores Choice, Score and Noul questions from embedding similarity and returns a distribution without generating text.
A Situated World Model for fast judgment in agent loops.
Decides
choice, score, noul, classify
choice, route, classify
Architecture
jevembed
trio-spark
Fine-tuned from
qwen/qwen3-embedding-4b
—
License
apache-2.0
proprietary
Availability
Open weights
Hosted API
Hosted by
—
MachineFi
Input price
—
$0.042/MTok
Decision accuracy
85.9%
—
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
p95 latency
—
—
Evaluation suite
JevEmbed-Data test split (64,110 hard-label questions), final LoRA checkpoint before merging
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 jevembed and trio-spark-v1?
jevembed is from HIT-TMG (Lychee Team) and trio-spark-v1 from MachineFi. jevembed has open weights you can download and run; trio-spark-v1 is only available as a hosted API. Both answer choice and classify questions. Only jevembed answers score and noul. Only trio-spark-v1 answers route. jevembed is licensed apache-2.0; trio-spark-v1, proprietary.
Which is more accurate, jevembed or trio-spark-v1?
Only jevembed publishes an accuracy figure (85.9% on JevEmbed-Data test split (64,110 hard-label questions), final LoRA checkpoint before merging); trio-spark-v1 does not, so there is no comparison to make without your own test.
Which is cheaper, jevembed or trio-spark-v1?
jevembed: Free (open weights). trio-spark-v1: $0.042 / $0 per 1M. Open weights cost nothing per call beyond your own hardware.
Can I run jevembed or trio-spark-v1 locally?
jevembed yes — systemone pull hit-tmg/jevembed downloads its weights. The other is only served as a hosted API.
