Safetensors
jev-style
English
decision-index
system-one
typed-decisions
rule-based
symbolic
no-weights
zero-parameters
program-execution
chess
bm25
tool-use
deterministic
cpu
Eval Results (legacy)
Instructions to use tardellirs/prumo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- jev-style
How to use tardellirs/prumo with jev-style:
pip install "jev-style[torch]"
from jev_style import JevStyle, noul, choice js = JevStyle.from_pretrained("tardellirs/prumo") out = js.decide("I was charged twice for one order.", { "billing": noul("This message is about billing."), "team": choice("Which team should handle it?", ["billing", "shipping", "tech"]), }) print(out["answers"]["team"]["choice"]) - Notebooks
- Google Colab
- Kaggle
Commit History
8 days ago
-
Link the paper (arXiv:2610.30970)
Verified signature5d7c9ef
11 days ago
-
Card: official Decision Index 0.4 Full score
Verified signatured8cec81
13 days ago
-
card: same text as the GitHub README, with the Hub metadata header
Verified signature6571fc0 -
card: same text as the GitHub README, with the Hub metadata header
Verified signatureb5865b9 -
card: same text as the GitHub README, with the Hub metadata header
Verified signature3cde1c6
14 days ago
-
Update README.md
Verified signature862bbd5 -
Rename constants.safetensors to model.safetensors so the Hub counts the 527 parameters (same bytes, sha256 088d42e0...)
Verified signaturef42441d -
Move every hand-set numeric parameter into constants.safetensors (527, F64); answers unchanged (public 0.3 = 20.14)
Verified signatured82d9c4 -
card: state the size as 700 hand-set constants (0 learned)
Verified signature2d5490a -
Prumo: zero-parameter rule-based decision engine (Decision Index 0.3, public 20.14)
Verified signatureb445c46 -
initial commit
Verified signaturec28afac