A 0.4B decision model made from the first 20 layers of Qwen3-0.6B plus a small attention head that compares options. It answers choice, noul and score questions in one pass, and choice order cannot change its answer by construction. Non-commercial licence.
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.
Decides
choice, score, noul
choice, score, noul
Architecture
bev-decider
lumma-fev
Fine-tuned from
qwen/qwen3-0.6b
frontiersmind/lumma-0.6b-base
License
cc-by-nc-4.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
74.7%
64.0%
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 bev-decider and lumma-fev?
bev-decider is from Avishek Biswas and lumma-fev from FrontiersMind. Both have open weights you can download and run. Both answer choice, score and noul questions. lumma-fev reads up to 8K tokens of state, against 2K tokens for bev-decider. bev-decider is the smaller model, at 478M parameters to 649M. bev-decider is licensed cc-by-nc-4.0; lumma-fev, apache-2.0.
Which is more accurate, bev-decider or lumma-fev?
They report on different suites — bev-decider 74.7% on avbiswas/bev-decision test split (5,000 held-out questions over 2,617 states), the maker's own, lumma-fev 64.0% on typed-decisions (maker's table; split not stated) — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, bev-decider or lumma-fev?
bev-decider: Free (open weights). lumma-fev: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run bev-decider or lumma-fev locally?
Yes, both: systemone pull avishek-biswas/bev-decider and systemone pull frontiersmind/lumma-fev download the weights.
Evaluation suite
avbiswas/bev-decision test split (5,000 held-out questions over 2,617 states), the maker's own
