Berget AI's System One model for gating agent commands: a LoRA adapter and a fine-tuned joint schema head on Cloudflare's Clef-Flash that answer noul, choice and score questions over a state in one forward pass. Trained on Swedish and English operations decisions.
Omar (kouhxp)'s CPU decision model: a Qwen3.5-0.8B fine-tune shipped as GGUF for llama.cpp that answers yes/no, choice and score questions with a probability per option plus a 'none of these' reject probability, served by a local Jev-style HTTP runtime.
Decides
noul, choice, score
choice, score, noul, classify, route
Architecture
clef
gutsy
Fine-tuned from
cloudflare/clef-flash
qwen/qwen3.5-0.8b
License
apache-2.0
apache-2.0
Availability
Open weights
Open weights
Hosted by
—
—
Input price
—
—
Decision accuracy
97.0%
73.2%
Calibration error
—
—
Valid action rate
—
—
Median latency
—
—
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 and gutsy?
bev is from Berget AI and gutsy from Omar (kouhxp). Both have open weights you can download and run. Both answer noul, choice and score questions. Only gutsy answers classify and route. gutsy is the smaller model, at 800M parameters to 9.0B.
Which is more accurate, bev or gutsy?
They report on different suites — bev 97.0% on Berget held-out risk split (16,902 questions; same operations-traffic corpora as training), gutsy 73.2% on JevBench public set (231 items; 169 correct), the maker's own run with Q8_0 — so the numbers do not rank them. Test both on your own labelled examples.
Which is cheaper, bev or gutsy?
bev: Free (open weights). gutsy: Free (open weights). Open weights cost nothing per call beyond your own hardware.
Can I run bev or gutsy locally?
Yes, both: systemone pull berget-ai/bev and systemone pull kouhxp/gutsy download the weights.
Evaluation suite
Berget held-out risk split (16,902 questions; same operations-traffic corpora as training)
JevBench public set (231 items; 169 correct), the maker's own run with Q8_0
