BasedAGIBasedAGI
Menu
Rankings live

business_productivity

Knowledge base Q&A (with citations)

Answer questions grounded in an internal KB, with evidence.

#1 Recommendation

gemini-3-pro-preview

Strong on BasedAGI KB Q&A Eval overall_score_pct (100%) and FACTS Benchmark Suite facts_grounding_score_pct (88%)

external/google/gemini-3-pro-preview

41.9%

Score

54.3%

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

BasedAGI KB Q&A Eval: overall_score_pct

All Ranked Models

Max params:
Min confidence:
30 of 30
RankModelScore
#1gemini-3-pro-preview

Strong on BasedAGI KB Q&A Eval overall_score_pct (100%) and FACTS Benchmark Suite facts_grounding_score_pct (88%)

41.9%
#2gemini-2.5-pro

Strong on FACTS Benchmark Suite facts_grounding_score_pct (100%) and BasedAGI KB Q&A Eval overall_score_pct (75%)

37.5%
#3anthropic/claude-sonnet-4.6

Strong on Vals Finance Agent overall_accuracy_pct (100%) and Vals Finance Agent complex_retrieval_accuracy_pct (91%)

31.4%
#4Grok-4-0709
27.5%
#5gpt-5-mini-2025-08-07
24.7%
#6gpt-5-2025-08-07
23.3%
#7google/gemini-3.1-pro-preview
22.3%
#8openai/gpt-5.4-2026-03-05
22.2%
#9gpt-4.1-20250414
22.1%
#10gemini-2.5-flash
20.3%
#11claude-opus-4-5-20251101
20.0%
#12claude-sonnet-4-20250514
19.3%
#13gpt-5.1-2025-11-13
19.2%
#14google/gemini-3.1-flash-lite-preview
18.8%
#15gemini-3-flash-preview
17.1%
#16xai-org/grok-4-fast-reasoning
17.1%
#17gpt-5.2-2025-12-11
16.7%
#18anthropic/claude-opus-4-6-thinking
16.3%
#19kimi/kimi-k2.5-thinking
16.1%
#20xai-org/grok-4-1-fast-reasoning
16.1%
#21anthropic/claude-opus-4-5-20251101-thinking
15.0%
#22anthropic/claude-sonnet-4-5-20250929-thinking
14.2%
#24mistralai/mistral-large-2512
13.1%
#26x-ai/grok-3
12.9%
#27zai/glm-5-thinking
12.8%
#28anthropic/claude-haiku-4-5-20251001-thinking
12.6%
#29xai-org/grok-4-1-fast-non-reasoning
12.5%
#30alibaba/qwen3.5-flash
12.4%
#31anthropic/claude-opus-4-1-20250805
12.3%
#32o3-20250416
11.6%

Compare Models

Model A leads by +4.4%

Shareable Link →

Model A

gemini-3-pro-preview

external/google/gemini-3-pro-preview

41.9%

Rank #1

Confidence 54.3%29 evidence pts

BasedAGI KB Q&A Eval: overall_score_pct

Value 100.0% · Conf 100.0% · Weight 4.2%

basedagi_kb_qna_eval.overall_score_pct (Mar 12, 2026)

FACTS Benchmark Suite: facts_grounding_score_pct

Value 88.3% · Conf 100.0% · Weight 3.4%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

Vals Finance Agent: overall_accuracy_pct

Value 87.0% · Conf 100.0% · Weight 3.1%

vals_finance_agent.overall_accuracy_pct (Mar 12, 2026)

FACTS Benchmark Suite: facts_search_score_pct

Value 100.0% · Conf 100.0% · Weight 3.0%

facts_benchmark_suite.facts_search_score_pct (Mar 12, 2026)

Model B

gemini-2.5-pro

external/google/gemini-2-5-pro

37.5%

Rank #2

Confidence 56.4%30 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 100.0% · Conf 100.0% · Weight 3.9%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

BasedAGI KB Q&A Eval: overall_score_pct

Value 75.0% · Conf 100.0% · Weight 3.2%

basedagi_kb_qna_eval.overall_score_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: overall_hallucination_error_pct

Value 76.0% · Conf 100.0% · Weight 3.0%

vectara_hhem_leaderboard.overall_hallucination_error_pct (Mar 12, 2026)

Vals Finance Agent: overall_accuracy_pct

Value 65.5% · Conf 100.0% · Weight 2.4%

vals_finance_agent.overall_accuracy_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 Mortgage Tax

vals_mortgage_tax

33 rows

1.1% avg lift

Vals Legal Bench

vals_legal_bench

32 rows

0.3% avg lift

Vals MedQA

vals_medqa

30 rows

0.3% 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.general_business

Related Use Cases