This directory contains bundled benchmark/eval datasets that can be loaded by:
/rlm import-evals pack=eval/packs/<file>/rlm bench preset=<name> pack=eval/packs/<file>
Included sample packs:
pydantic_time_range_v1.yaml(dataset-stylecases)google_adk_memory_eval.json(Google ADK-styleeval_cases)superoptix_qa_pairs.json(generic JSON records)rlm_x_claims_matrix.yaml(RLM-vs-harness claims validation matrix)