BasedAGIBasedAGI
Menu
Rankings live

geopolitics

Narrative tracking

Track narratives across multi-lingual sources and flag contradictions.

#1 Recommendation

gemini-3-pro-preview

Strong on FACTS Benchmark Suite facts_grounding_score_pct (88%) and FACTS Benchmark Suite facts_search_score_pct (100%)

external/google/gemini-3-pro-preview

28.9%

Score

37.7%

Confidence

Runners-up:#2 gemini-2.5-pro (25.8%)#3 gemini-2.5-flash (24.1%)#4 anthropic/claude-sonnet-4.6 (21.7%)

Limited benchmark evidence for this use case.

53 ranked models with average evidence of 14.1 points. Rankings may shift as more benchmark data is ingested.

Ranked Models

30

Evidence Quality

82%

Scoring

Benchmark-backed

Top Signal

FACTS Benchmark Suite: facts_grounding_score_pct

All Ranked Models

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

Strong on FACTS Benchmark Suite facts_grounding_score_pct (88%) and FACTS Benchmark Suite facts_search_score_pct (100%)

28.9%
#2gemini-2.5-pro

Strong on FACTS Benchmark Suite facts_grounding_score_pct (100%) and Vectara HHEM Leaderboard overall_hallucination_error_pct (76%)

25.8%
#3gemini-2.5-flash

Strong on LanguageBench overall:mean (100%) and FACTS Benchmark Suite facts_grounding_score_pct (90%)

24.1%
#4anthropic/claude-sonnet-4.6
21.7%
#5gpt-5-mini-2025-08-07
21.3%
#6gpt-4.1-20250414
21.1%
#7gpt-5-2025-08-07
20.9%
#8Grok-4-0709
20.6%
#9google/gemini-3.1-pro-preview
20.0%
#10openai/gpt-5.4-2026-03-05
19.5%
#11claude-opus-4-5-20251101
18.9%
#12gpt-5.1-2025-11-13
17.4%
#13claude-sonnet-4-20250514
17.3%
#14gemini-3-flash-preview
16.9%
#15google/gemini-3.1-flash-lite-preview
16.2%
#16xai-org/grok-4-fast-reasoning
16.0%
#17xai-org/grok-4-1-fast-reasoning
15.3%
#18anthropic/claude-opus-4-6-thinking
15.2%
#19gpt-5.2-2025-12-11
15.2%
#22kimi/kimi-k2.5-thinking
14.4%
#23anthropic/claude-opus-4-5-20251101-thinking
14.0%
#24anthropic/claude-sonnet-4-5-20250929-thinking
13.2%
#27zai/glm-5-thinking
12.8%
#29x-ai/grok-3
12.7%
#30google/gemini-2.0-flash-001
12.6%
#31o3-20250416
11.9%
#32mistralai/mistral-large-2512
11.9%
#34anthropic/claude-haiku-4-5-20251001-thinking
11.7%
#35alibaba/qwen3.5-flash
11.5%
#36xai-org/grok-4-1-fast-non-reasoning
11.5%

Compare Models

Model A leads by +3.2%

Shareable Link →

Model A

gemini-3-pro-preview

external/google/gemini-3-pro-preview

28.9%

Rank #1

Confidence 37.7%23 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 88.3% · Conf 100.0% · Weight 2.8%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

FACTS Benchmark Suite: facts_search_score_pct

Value 100.0% · Conf 100.0% · Weight 2.4%

facts_benchmark_suite.facts_search_score_pct (Mar 12, 2026)

FACTS Benchmark Suite: average_score_pct

Value 100.0% · Conf 100.0% · Weight 2.2%

facts_benchmark_suite.average_score_pct (Mar 12, 2026)

Vals Finance Agent: overall_accuracy_pct

Value 87.0% · Conf 100.0% · Weight 2.2%

vals_finance_agent.overall_accuracy_pct (Mar 12, 2026)

Model B

gemini-2.5-pro

external/google/gemini-2-5-pro

25.8%

Rank #2

Confidence 46.3%25 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 100.0% · Conf 100.0% · Weight 3.1%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: overall_hallucination_error_pct

Value 76.0% · Conf 100.0% · Weight 2.5%

vectara_hhem_leaderboard.overall_hallucination_error_pct (Mar 12, 2026)

Vals CorpFin v2: overall_accuracy_pct

Value 78.4% · Conf 100.0% · Weight 2.0%

vals_corp_fin_v2.overall_accuracy_pct (Mar 12, 2026)

FACTS Benchmark Suite: average_score_pct

Value 78.3% · Conf 100.0% · Weight 1.7%

facts_benchmark_suite.average_score_pct (Mar 12, 2026)

▶Ranking Diagnostics & Missing Models

Source Lift

Ranked

53

Sources

8

Quality

Insufficient

Vals CorpFin v2

vals_corp_fin_v2

42 rows

1.6% avg lift

Vals Legal Bench

vals_legal_bench

38 rows

0.4% avg lift

Vals Tax Eval v2

vals_tax_eval_v2

37 rows

0.4% avg lift

Vals MedQA

vals_medqa

36 rows

0.4% avg lift

Missing Strong Models

gpt-4o

external/openai/gpt-4o

Rank #22

15.2%

Thin evidence after weighting

deepseek-v3

external/deepseek-ai/deepseek-v3

Rank #66

8.8%

Thin evidence after weighting
▶Taxonomy Details

Core Tasks

task.multi_doc_synthesistask.contradiction_detection

Required Modes

mode.long_context

Domains

domain.geopolitics_intel

Related Use Cases

geopolitics

Political risk brief

Summarize key developments into risks, scenarios, and actions.

Top: gemini-3-pro-preview