BasedAGIBasedAGI
Menu
Rankings live

healthcare

Patient education bot (RAG grounded)

Answer patient FAQ using trusted sources with cautious wording.

#1 Recommendation

gemini-2.5-pro

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

external/google/gemini-2-5-pro

27.2%

Score

51.1%

Confidence

Limited benchmark evidence for this use case.

53 ranked models with average evidence of 16.2 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-2.5-pro

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

27.2%
#2gemini-2.5-flash

Strong on FACTS Benchmark Suite facts_grounding_score_pct (90%) and LanguageBench Translation Official (Split) translation_to:bleu (92%)

27.1%
#3gemini-3-pro-preview

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

24.9%
#4gpt-4.1-20250414
24.7%
#5gpt-5-mini-2025-08-07
21.2%
#6claude-sonnet-4-20250514
20.4%
#7gpt-5-2025-08-07
19.9%
#8Grok-4-0709
19.5%
#9google/gemini-3.1-pro-preview
19.0%
#10claude-opus-4-5-20251101
18.7%
#11openai/gpt-5.4-2026-03-05
18.1%
#13anthropic/claude-sonnet-4.6
17.8%
#14gemini-3-flash-preview
16.5%
#16gpt-5.1-2025-11-13
16.1%
#17gpt-4.1-mini-20250414
15.3%
#18xai-org/grok-4-fast-reasoning
15.2%
#19xai-org/grok-4-1-fast-reasoning
14.2%
#20anthropic/claude-opus-4-6-thinking
14.1%
#22gpt-5.2-2025-12-11
13.9%
#23google/gemini-2.0-flash-001
13.9%
#24google/gemini-3.1-flash-lite-preview
13.5%
#25anthropic/claude-opus-4-5-20251101-thinking
13.3%
#26kimi/kimi-k2.5-thinking
12.7%
#27anthropic/claude-sonnet-4-5-20250929-thinking
12.5%
#28anthropic/claude-opus-4-1-20250805
12.3%
#29gpt-4o
12.1%
#31x-ai/grok-3
11.7%
#32o3-20250416
11.6%
#33xai-org/grok-4-1-fast-non-reasoning
11.4%
#34qwen-2.5-72b-instruct
11.2%

Compare Models

Model A leads by +0.1%

Shareable Link →

Model A

gemini-2.5-pro

external/google/gemini-2-5-pro

27.2%

Rank #1

Confidence 51.1%34 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 1.8%

vectara_hhem_leaderboard.overall_hallucination_error_pct (Mar 12, 2026)

Vals CorpFin v2: overall_accuracy_pct

Value 78.4% · Conf 100.0% · Weight 1.4%

vals_corp_fin_v2.overall_accuracy_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: medicine_hallucination_error_pct

Value 92.8% · Conf 100.0% · Weight 1.4%

vectara_hhem_leaderboard.medicine_hallucination_error_pct (Mar 12, 2026)

Model B

gemini-2.5-flash

external/google/gemini-2-5-flash

27.1%

Rank #2

Confidence 36.1%22 evidence pts

FACTS Benchmark Suite: facts_grounding_score_pct

Value 90.4% · Conf 100.0% · Weight 2.8%

facts_benchmark_suite.facts_grounding_score_pct (Mar 12, 2026)

LanguageBench Translation Official (Split): translation_to:bleu

Value 92.0% · Conf 100.0% · Weight 2.7%

languagebench_translation_official.translation_to_bleu (Mar 12, 2026)

BRIDGE Medical Leaderboard: average_performance_pct

Value 100.0% · Conf 100.0% · Weight 2.3%

bridge_medical_leaderboard.average_performance_pct (Mar 12, 2026)

Vectara HHEM Leaderboard: overall_hallucination_error_pct

Value 72.4% · Conf 100.0% · Weight 1.7%

vectara_hhem_leaderboard.overall_hallucination_error_pct (Mar 12, 2026)

Ranking Diagnostics & Missing Models

Source Lift

Ranked

53

Sources

8

Quality

Insufficient

Vals CorpFin v2

vals_corp_fin_v2

40 rows

1.1% avg lift

Vals MedQA

vals_medqa

39 rows

1.9% avg lift

Vals Finance Agent

vals_finance_agent

29 rows

1.1% avg lift

Vals MedScribe

vals_medscribe

28 rows

1.3% avg lift

Missing Strong Models

gpt-4o-20241120

external/openai/gpt-4o-20241120

Rank #49

10.7%

Thin evidence after weighting
Taxonomy Details

Core Tasks

task.rag_answer_with_citationstask.rewrite_clarity

Required Modes

mode.citations

Domains

domain.healthcare_clinical

Related Use Cases