Compare MTP steps and prefill batches; retain qualified MTP 2 baseline

This commit is contained in:
2026-09-18 23:22:45 +08:00
parent 918b78a3b8
commit c8bfe6b9ab
45 changed files with 3785 additions and 3 deletions
@@ -0,0 +1,8 @@
{"name": "invoice-extraction", "passed": true, "elapsed_s": 1.1935799049970228}
{"name": "deduplicate", "passed": true, "elapsed_s": 1.5906357709973236}
{"name": "schedule", "passed": true, "elapsed_s": 1.9658004070006427}
{"name": "chinese-policy", "passed": true, "elapsed_s": 1.2027164360042661}
{"name": "merge-intervals", "passed": true, "validation_stdout": "passed\n", "validation_stderr": ""}
{"name": "binary-search", "passed": true, "validation_stdout": "passed\n", "validation_stderr": ""}
{"name": "tool-arguments", "passed": true}
QUALITY_COMPLETE