refactor(benchmarks): modularize benchmark system with config-driven execution

This commit is contained in:
2026-03-02 16:00:36 +08:00
parent a7b01cb49e
commit a16b376dd7
14 changed files with 779 additions and 180 deletions

View File

@@ -0,0 +1,6 @@
"""Benchmark evaluation tasks."""
from .retrieval import RetrievalTask
from .registry import TASK_REGISTRY, get_task
__all__ = ["RetrievalTask", "TASK_REGISTRY", "get_task"]