rlcd-modernbert-151m

What it is

At 151M parameters this is the smallest entry in the directory, built on GLiClass over a ModernBERT encoder. GLiClass does zero-shot classification against label sets supplied at inference time, which maps onto the choice primitive almost directly: the decision space is an argument, not something baked into a trained head.

The name is a claim about method

RLCD appears in the model’s name and is also the training method TypeSafe describes for Jev. The name is the only evidence we have of a connection. We have not verified that this model was trained the same way, and a shared acronym is not a shared recipe.

Why the size is the interesting part

An encoder this small runs on a CPU. No GPU, no per-token bill, no external API in the path of a decision. For the high-volume, low-stakes end of a decision workload, that is a different economic proposition to anything else on this list, and it only needs to be good enough, not best.

The failure mode to look for is overconfidence. Small classifiers tend to push probability mass to the edges, which inflates accuracy-at-threshold while wrecking calibration. Accuracy alone will not catch it; expected calibration error will.

Status

Not yet benchmarked by us. It is cheap to evaluate, minutes on CPU for the full 400-case split, so it will run in the first open-weight batch alongside the other encoders.