RT-P (rt-p)

RT-P is a Relational Transformer checkpoint pair for in-context / few-shot entity prediction over multi-table relational databases (no per-task gradient training). Same architecture and file layout as stanford-star/rt-j — drop-in replacement.

Variant	Folder	Task
Classifier	classification/	binary entity classification
Regressor	regression/	entity regression

Architecture: ~85.6M params · bfloat16 · 12 blocks, d_model 512, 8 heads, d_ff 2048 · text columns embedded with all-MiniLM-L12-v2 (d_text 384). Each folder contains model.safetensors (weights) and config.json (dims + text-embedding model).

Usage

from rt.model import RelationalTransformer

model = RelationalTransformer.from_pretrained("stanford-star/rt-p/classification")
# or: RelationalTransformer.from_pretrained("stanford-star/rt-p", subfolder="regression")

pixi run eval --model.load-ckpt-path stanford-star/rt-p/classification ...

Related

• Training data: stanford-star/the-join
• Evaluation: RelBench
• Reference model: stanford-star/rt-j