wexumin/xrag-7b-router

xRAG Router

A routing classifier for xRAG that decides at inference time whether compressed context is sufficient or whether to fall back to the full context.

Trained with overflowguard on SQuAD using an LLM judge (DeepSeek) for evaluation, with 5-fold stratified CV for threshold selection (Youden's J statistic).

Usage

from examples.train_xrag import XragRouter

# Load base model + routing classifier in one call
router = XragRouter.from_pretrained("wexumin/xrag-7b-router")
router.park_gpu()

# Auto-routing: classifier decides compressed vs full
result = router.run_pipeline("long document text...", query="what is X?")
print(result["prediction"])  # answer
print(result["mode"])        # "compressed" or "full"
print(result["tokens_saved"])

What's in this repo

• router_config.json — base model path + routing threshold
• routing_clf.pt — classifier weights (2-layer MLP, ~2K params)

The base xRAG model is loaded automatically from Hannibal046/xrag-7b. This repo only contains the lightweight routing classifier on top. This model also requires custom code from xRAG repository.