LLM Hallucination Detection - Non-RAG Leaderboard — The Aggregate

LLM Hallucination Detection - Non-RAG — leaderboard

Metric: Hallucination rate (%). Source: huggingface.co. 17 models tracked.

Top models

#	Model	Score
1	Gemini 2.5 Pro	0.36
2	Gemma 3 27B (IT)	0.48
3	Claude Sonnet 4	0.6
4	GPT-4o	0.64
5	Llama 4 Maverick Instruct FP8	0.69
6	DeepSeek R1 0528	0.78
7	Qwen 3 235B A22B 2507	0.82
8	DeepSeek V3 (0324)	0.91
9	Qwen 3 235B A22B FP8	2.09
10	Llama 4 Scout Instruct	2.48
11	Kimi K2	3.04
12	Qwen 2.5 VL 7B Instruct	4.55
13	Mistral Small 3	7.85
14	Mistral Nemo Instruct (2407)	8.74
15	Magistral Small	28.07

Interactive version: theaggregate.ai/benchmark?slug=llm-hallucination-detection-non-rag · How the rankings work · Data refreshed daily, snapshot 2026-07-22.