Ranking

Best AI models for Factuality & hallucination

Answering factual questions correctly, abstaining when unsure, and staying grounded in provided sources.

Ranked models

Based on 1 benchmark
Factuality & hallucination ranking
RankModelCompanyScoreCoverageReleased
1GPT-6 AstraOpenAI75.61 of 13 Sep 2026
2Gemini 3.1 ProGoogle DeepMind73.51 of 119 Feb 2026
3Claude Opus 5.5Anthropic72.21 of 122 Sep 2026
4Claude Fable 5.1Anthropic70.81 of 11 Sep 2026
5Claude Fable 5Anthropic70.71 of 19 Jun 2026
6Gemini 3.8 FlashGoogle DeepMind69.71 of 12 Sep 2026
6GPT-5.6 SolOpenAI69.71 of 19 Jul 2026
8Gemini 3.7 FlashGoogle DeepMind69.21 of 113 Aug 2026
9Gemini 3 FlashGoogle DeepMind66.81 of 117 Dec 2025
10Gemini 3.5 FlashGoogle DeepMind66.21 of 119 May 2026
10Gemini 3.6 FlashGoogle DeepMind66.21 of 121 Jul 2026
12GPT-5.5OpenAI63.01 of 123 Apr 2026
13Muse Spark 1.2Meta60.31 of 15 Aug 2026
14Claude Opus 5Anthropic59.91 of 124 Jul 2026
15Muse Spark 1.1Meta57.81 of 18 Apr 2026
16Qwen3.7-MaxAlibaba Group55.81 of 120 May 2026
17Claude Opus 4.8Anthropic53.01 of 128 May 2026
18DeepSeek-V4-ProDeepSeek52.91 of 124 Apr 2026
19Qwen3.6-Max-Preview DeprecatedAlibaba Group52.01 of 114 Apr 2026
20Claude Opus 4.7Anthropic51.71 of 116 Apr 2026
21Kimi K3Moonshot AI50.61 of 116 Jul 2026
22GPT-5 DeprecatedOpenAI50.11 of 17 Aug 2025
23o3 DeprecatedOpenAI49.41 of 116 Apr 2025
24Grok 4.6SpaceXAI (xAI)49.31 of 112 Aug 2026
25Qwen3-Max DeprecatedAlibaba Group48.71 of 15 Sep 2025
26Grok 4.5SpaceXAI (xAI)48.31 of 18 Jul 2026
27GPT-5.1OpenAI48.01 of 112 Nov 2025
28Qwen3.8-MaxAlibaba Group47.31 of 12 Aug 2026
29Claude Opus 4.6Anthropic47.01 of 15 Feb 2026
30GPT-5.4 ProOpenAI46.31 of 15 Mar 2026
31Claude Opus 4.5Anthropic45.71 of 124 Nov 2025
32GPT-5.4OpenAI45.11 of 15 Mar 2026
33Qwen3.6-PlusAlibaba Group44.11 of 12 Apr 2026
34GPT-5.6 TerraOpenAI43.21 of 19 Jul 2026
35o1 DeprecatedOpenAI41.11 of 15 Dec 2024
36GLM-5.3Z.ai (Zhipu AI)41.01 of 118 Aug 2026
36GPT-5.6 LunaOpenAI41.01 of 19 Jul 2026
38Qwen3-235B-A22B-Thinking-2507Alibaba Group40.41 of 125 Jul 2025
39InklingThinking Machines Lab40.31 of 115 Jul 2026
40GPT-5.2OpenAI37.11 of 111 Dec 2025
41Kimi K2.7 CodeMoonshot AI36.51 of 112 Jun 2026
42Claude Sonnet 4.6Anthropic35.51 of 117 Feb 2026
43Kimi K2.6Moonshot AI34.91 of 120 Apr 2026
44Kimi K2.5Moonshot AI34.31 of 127 Jan 2026
45GLM-5.2Z.ai (Zhipu AI)34.21 of 116 Jun 2026
46GLM-5.1Z.ai (Zhipu AI)34.01 of 17 Apr 2026
47Claude Sonnet 5Anthropic33.71 of 130 Jun 2026
48DeepSeek-V4-FlashDeepSeek33.61 of 124 Apr 2026
49Grok 4.3SpaceXAI (xAI)33.21 of 117 Apr 2026
50GLM-4.7Z.ai (Zhipu AI)32.21 of 122 Dec 2025
51GPT-4.1OpenAI31.11 of 114 Apr 2025
52Claude Sonnet 4.5Anthropic30.71 of 129 Sep 2025
53Grok 4.20SpaceXAI (xAI)30.21 of 1Feb 2026
54GPT-5.4 miniOpenAI29.41 of 117 Mar 2026
55GPT-4oOpenAI26.01 of 113 May 2024
56Qwen3.5-PlusAlibaba Group25.41 of 115 Feb 2026
57GPT-5 mini DeprecatedOpenAI21.61 of 17 Aug 2025
58Qwen3.5-FlashAlibaba Group20.31 of 123 Feb 2026
59o4-mini DeprecatedOpenAI19.61 of 116 Apr 2025
60Inkling SmallThinking Machines Lab19.11 of 130 Jul 2026
61Qwen3.6-FlashAlibaba Group15.91 of 116 Apr 2026
62o3-mini DeprecatedOpenAI15.31 of 131 Jan 2025
63Claude Haiku 4.5Anthropic13.21 of 115 Oct 2025
64GPT-4.1 miniOpenAI12.71 of 114 Apr 2025
65Claude 3 Opus RetiredAnthropic12.61 of 14 Mar 2024
66GPT-5 nano DeprecatedOpenAI11.71 of 17 Aug 2025
66GPT-5.4 nanoOpenAI11.71 of 117 Mar 2026
68Gemma 4 31BGoogle DeepMind10.41 of 131 Mar 2026
69GPT-4o miniOpenAI8.31 of 118 Jul 2024
70GPT-4.1 nano DeprecatedOpenAI6.01 of 114 Apr 2025
Best AI models for Factuality & hallucination (September 2026) · supermarket.dev