BasedAGIBasedAGI
Menu
Rankings live

legal

Contract Q&A (RAG grounded)

Answer contract questions grounded in the actual contract text.

#1 Recommendation

gemini-2.5-pro

Strong on FACTS Benchmark Suite facts_grounding_score_pct (100%) and LEXam Leaderboard average_score_pct (89%)

external/google/gemini-2-5-pro

33.3%

Score

52.0%

Confidence

Limited benchmark evidence for this use case.

50 ranked models with average evidence of 15.9 points. Rankings may shift as more benchmark data is ingested.

Ranked Models

30

Evidence Quality

84%

Scoring

Benchmark-backed

Top Signal

FACTS Benchmark Suite: facts_grounding_score_pct

All Ranked Models

Max params:
Min confidence:
30 of 30
RankModelScore
#1gemini-2.5-pro

Strong on FACTS Benchmark Suite facts_grounding_score_pct (100%) and LEXam Leaderboard average_score_pct (89%)

33.3%
#2gemini-3-pro-preview

Strong on FACTS Benchmark Suite facts_grounding_score_pct (88%) and Vals Legal Bench overall_accuracy_pct (99%)

33.1%
#3gpt-4.1-20250414

Strong on Vals Case Law v2 overall_accuracy_pct (86%) and Vectara HHEM Leaderboard overall_hallucination_error_pct (82%)

26.5%
#4gpt-5-mini-2025-08-07
25.4%
#5anthropic/claude-sonnet-4.6
25.4%
#6gemini-2.5-flash
25.0%
#7google/gemini-3.1-pro-preview
23.4%
#8openai/gpt-5.4-2026-03-05
23.1%
#9Grok-4-0709
23.1%
#10gpt-5-2025-08-07
22.5%
#11claude-sonnet-4-20250514
21.5%
#12claude-opus-4-5-20251101
21.3%
#13gpt-5.1-2025-11-13
19.9%
#14gemini-3-flash-preview
19.7%
#15xai-org/grok-4-fast-reasoning
19.7%
#16google/gemini-3.1-flash-lite-preview
19.1%
#17xai-org/grok-4-1-fast-reasoning
18.6%
#18gpt-5.2-2025-12-11
16.5%
#19anthropic/claude-opus-4-6-thinking
16.2%
#20mistralai/mistral-large-2512
15.9%
#21anthropic/claude-opus-4-5-20251101-thinking
15.3%
#22x-ai/grok-3
15.3%
#23anthropic/claude-sonnet-4-5-20250929-thinking
14.6%
#24anthropic/claude-opus-4-1-20250805
14.3%
#28xai-org/grok-4-1-fast-non-reasoning
13.8%
#29zai/glm-5-thinking
13.7%
#30Command A (03-2025)
13.5%
#31kimi/kimi-k2.5-thinking
13.3%
#33Kimi K2 Thinking
12.9%
#34anthropic/claude-haiku-4-5-20251001-thinking
12.5%

Compare Models

Model A leads by +0.2%

Shareable Link →

Model A

gemini-2.5-pro

external/google/gemini-2-5-pro

33.3%

Rank #1

Confidence 52.0%30 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 100.0% · Conf 100.0% · Weight 3.5%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

LEXam Leaderboard: average_score_pct

Value 89.4% · Conf 100.0% · Weight 2.1%

lexam_leaderboard.average_score_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: overall_hallucination_error_pct

Value 76.0% · Conf 100.0% · Weight 2.0%

vectara_hhem_leaderboard.overall_hallucination_error_pct (Mar 12, 2026)

Galileo Agent Leaderboard v2: Avg TSQ

Value 79.5% · Conf 100.0% · Weight 1.8%

galileo_agent_v2.avg_tsq (Mar 12, 2026)

Model B

gemini-3-pro-preview

external/google/gemini-3-pro-preview

33.1%

Rank #2

Confidence 44.4%26 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 88.3% · Conf 100.0% · Weight 3.1%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

Vals Legal Bench: overall_accuracy_pct

Value 99.2% · Conf 100.0% · Weight 2.3%

vals_legal_bench.overall_accuracy_pct (Mar 12, 2026)

FACTS Benchmark Suite: facts_search_score_pct

Value 100.0% · Conf 100.0% · Weight 1.9%

facts_benchmark_suite.facts_search_score_pct (Mar 12, 2026)

LEXam Leaderboard: average_score_pct

Value 75.9% · Conf 100.0% · Weight 1.8%

lexam_leaderboard.average_score_pct (Mar 12, 2026)

Ranking Diagnostics & Missing Models

Source Lift

Ranked

50

Sources

8

Quality

Insufficient

Vals CorpFin v2

vals_corp_fin_v2

42 rows

1.3% avg lift

Vals Legal Bench

vals_legal_bench

41 rows

2.0% avg lift

Vals MedQA

vals_medqa

32 rows

0.3% avg lift

Vals Finance Agent

vals_finance_agent

29 rows

1.2% avg lift

Missing Strong Models

gpt-4o

external/openai/gpt-4o

Rank #22

15.2%

Thin evidence after weighting
Taxonomy Details

Core Tasks

task.rag_answer_with_citationstask.kb_navigation

Required Modes

mode.citations

Domains

domain.legal_contracts

Related Use Cases