﻿run_id	label	prompt_id	matter_id	thinking_mode	status	total_criteria	passed_criteria	score	strict_pass	failed_criterion_ids	total_ms	llm_ms	tool_ms	overhead_ms	tool_count	llm_call_count
20260908_develop	9/8develop(Gemini3.8)	GDPR_正当な利益_標準	69	standard	success	1	1	1.0000	True		28221	25269	791	2161	1	2
20260908_develop	9/8develop(Gemini3.8)	GDPR_注意点_標準	69	standard	success	1	1	1.0000	True		28144	27153	0	991	0	1
20260908_develop	9/8develop(Gemini3.8)	sヒアリング事項_検討_標準	67	standard	success	1	1	1.0000	True		32152	29078	1443	1631	6	2
20260908_develop	9/8develop(Gemini3.8)	s法律相談_初期分析_思考	69	thinking	success	3	3	1.0000	True		160941	152726	3904	4311	16	17
20260908_develop	9/8develop(Gemini3.8)	s法律相談_初期分析_標準	69	standard	success	3	3	1.0000	True		68375	62890	2877	2608	12	7
20260908_develop	9/8develop(Gemini3.8)	s法律相談_初期分析_高速	69	fast	success	3	3	1.0000	True		48199	41707	1951	4541	8	7
20260908_develop	9/8develop(Gemini3.8)	s過去案件_参考探索_標準	67	standard	success	1	1	1.0000	True		22112	19951	100	2061	1	2
20260908_develop	9/8develop(Gemini3.8)	ヒアリング_ロック社参考_標準	67	standard	success	1	1	1.0000	True		24147	22685	86	1376	1	2
20260908_develop	9/8develop(Gemini3.8)	ヒアリング案_メール文案_標準	67	standard	success	3	3	1.0000	True		12059	9928	0	2131	0	1
20260908_develop	9/8develop(Gemini3.8)	ヒアリング案_作成_標準	67	standard	success	1	1	1.0000	True		14067	12389	0	1678	0	1
20260908_develop	9/8develop(Gemini3.8)	法律相談_ヒアリング案_標準	69	standard	success	1	1	1.0000	True		14066	11730	0	2336	0	1
20260908_develop	9/8develop(Gemini3.8)	法律相談_英語化_標準	69	standard	success	1	1	1.0000	True		10056	8829	0	1227	0	1
20260908_develop	9/8develop(Gemini3.8)	論点整理_解決経緯_標準	63	standard	success	2	2	1.0000	True		40195	34004	2870	3321	6	3
20260908_develop	9/8develop(Gemini3.8)	過去案件_ロック社LLM_標準	67	standard	success	1	1	1.0000	True		14066	12794	92	1180	1	2
20260908_develop	9/8develop(Gemini3.8)	ALL		ALL		23	23	1.0000	1.0000		36914	33652	1008	2254	4	4
