jevhome/jevhome-L

jevhome L

A typed-decision model (cross-encoder, ~396M parameters, fp32) for jevhome: give it a state and yes/no (noul), pick-one (choice) or graded (score) questions, and it returns calibrated probabilities in one forward pass, on a plain CPU. Same API as Jev (POST /v1/systemone).

The order of the options of a choice question does not change the answer: all options get the same position ids, so the network reads them as a set, not a sequence, and a permutation gives the same probabilities up to floating-point rounding (int8 rounding can rarely change a close call).

pytorch/ holds the PyTorch weights (backbone model.safetensors + decision head head.safetensors) for GPU or research use; jevhome only needs the ONNX files (--exclude 'pytorch/*' skips them).

Run it

git clone https://github.com/Jibril-Frej/jev-at-home && cd jev-at-home && cargo build --release    # the jevhome binary (needs Rust)
hf download jevhome/jevhome-L --local-dir models/jevhome-L --exclude 'pytorch/*'
./target/release/jevhome serve models/jevhome-L

Numbers

JevBench Easy	Standard	Hard	ECE (Easy+Standard, lower is better)	External mean (10 sets)	Latency p50	p90	Peak RAM
100	93	38	0.126	68.9	195 ms	561 ms	2.1 GB

Accuracy in %, measured with the jevhome binary. Latency: one decision at a time, 4 threads on 4 cores of an AMD EPYC 9654. The full tables (every set, Jev, Laya and the teacher), the training recipe and the data are in the jevhome README.

Files

• model.onnx: the network, probabilities come from calibrated logits.
• decision_config.json, tokenizer*.json, config.json: input format and tokenizer.
• calibration_u.json: one temperature per question type and number of options.

Licence

CC BY-NC 4.0 (non-commercial): several training sources are non-commercial or share-alike. Backbone: answerdotai/ModernBERT-Large-Instruct.