FrontiersMind's decision model on a base it pre-trained from scratch. Each question reads the state and its own tokens once and a 256-dimension pointer head scores the options; Choice and Score take up to 255 entries, and nothing is generated.
TokenRhythm's prefill-only decision model for agent workflows. Built on its NeoHorse-1-4B (a Qwen3.5-4B derivative), it predicts Choice, Noul and Score answers over application-defined options without generating text, with optional single-image input.
Decides
choice, score, noul
choice, score, noul, classify, route
Architecture
lumma-fev
neohorse
Fine-tuned from
frontiersmind/lumma-0.6b-base
tokenrhythm/neohorse-1-4b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
64.0%
75.3%
Calibration error
—
—
Valid action rate
—
—
Median latency
45.8 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 lumma-fev and neohorse-jev?
lumma-fev is from FrontiersMind and neohorse-jev from TokenRhythm. Both have open weights you can download and run. Both answer choice, score and noul questions. Only neohorse-jev answers classify and route. lumma-fev is the smaller model, at 649M parameters to 4.0B.
Which is more accurate, lumma-fev or neohorse-jev?
They report on different suites — lumma-fev 64.0% on typed-decisions (maker's table; split not stated), neohorse-jev 75.3% on JevBench public set (231 items), vLLM, maker's run — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, lumma-fev or neohorse-jev?
lumma-fev: Free (open weights). neohorse-jev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run lumma-fev or neohorse-jev locally?
Yes, both: systemone pull frontiersmind/lumma-fev and systemone pull tokenrhythm/neohorse-jev download the weights.
—
Evaluation suite
typed-decisions (maker's table; split not stated)
JevBench public set (231 items), vLLM, maker's run
