BasedAGIBasedAGI
Menu
Rankings live

risk_eval

Disinformation and manipulation resistance (eval)

Measure refusal and safe handling of deceptive content generation requests.

#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

32.8%

Score

42.7%

Confidence

Limited benchmark evidence for this use case.

56 ranked models with average evidence of 13.6 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-3-pro-preview

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

32.8%
#2gemini-2.5-pro

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

30.2%
#3Grok-4-0709

Strong on Vals CorpFin v2 overall_accuracy_pct (94%) and Vals Finance Agent overall_accuracy_pct (84%)

24.8%
#4anthropic/claude-sonnet-4.6
24.6%
#5gpt-5-mini-2025-08-07
24.1%
#6gpt-4.1-20250414
23.7%
#7gpt-5-2025-08-07
23.7%
#8google/gemini-3.1-pro-preview
22.6%
#9openai/gpt-5.4-2026-03-05
22.1%
#10claude-opus-4-5-20251101
21.5%
#11xai-org/grok-4-fast-reasoning
19.7%
#12gpt-5.1-2025-11-13
19.7%
#13claude-sonnet-4-20250514
19.6%
#14xai-org/grok-4-1-fast-reasoning
19.2%
#15gemini-3-flash-preview
19.1%
#16google/gemini-3.1-flash-lite-preview
18.4%
#17anthropic/claude-opus-4-6-thinking
17.2%
#18gpt-5.2-2025-12-11
17.2%
#19gemini-2.5-flash
16.9%
#20kimi/kimi-k2.5-thinking
16.3%
#21x-ai/grok-3
16.1%
#22anthropic/claude-opus-4-5-20251101-thinking
15.9%
#23anthropic/claude-sonnet-4-5-20250929-thinking
14.9%
#25zai/glm-5-thinking
14.5%
#26gpt-4o-2024-05-13
14.4%
#27Llama-2-7b-chat-hf
14.1%
#28xai-org/grok-4-1-fast-non-reasoning
13.8%
#30openai/gpt-4o-mini-2024-07-18
13.7%
#31o3-20250416
13.5%
#32mistralai/mistral-large-2512
13.5%

Compare Models

Model A leads by +2.6%

Shareable Link →

Model A

gemini-3-pro-preview

external/google/gemini-3-pro-preview

32.8%

Rank #1

Confidence 42.7%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 2.3%

facts_benchmark_suite.average_score_pct (Mar 12, 2026)

Vals Finance Agent: overall_accuracy_pct

Value 87.0% · Conf 100.0% · Weight 2.3%

vals_finance_agent.overall_accuracy_pct (Mar 12, 2026)

Model B

gemini-2.5-pro

external/google/gemini-2-5-pro

30.2%

Rank #2

Confidence 43.7%24 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 100.0% · Conf 100.0% · Weight 3.2%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: overall_hallucination_error_pct

Value 76.0% · Conf 100.0% · Weight 2.6%

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 1.8%

facts_benchmark_suite.average_score_pct (Mar 12, 2026)

Ranking Diagnostics & Missing Models

Source Lift

Ranked

56

Sources

8

Quality

Insufficient

Vals CorpFin v2

vals_corp_fin_v2

42 rows

1.6% avg lift

Vals Legal Bench

vals_legal_bench

37 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/deepseek-r1

external/deepseek/deepseek-r1

Rank #54

10.5%

Thin evidence after weighting
Taxonomy Details

Core Tasks

task.refusal_rate_by_categorytask.jailbreak_resistance

Required Modes

none

Domains

domain.general_businessdomain.geopolitics_intel

Related Use Cases