AA-Omniscience Index leaderboard

A display-only Artificial Analysis factual knowledge index.

Models ranked by their published Artificial Analysis Omniscience Index value

Rank	Model	Provider	Index score
1	Claude Fable 5.1	Anthropic	43.5
2	GPT-6 Astra	OpenAI	43.4
3	Claude Fable 5	Anthropic	43.3
4	Claude Opus 5	Anthropic	37.1
5	Gemini 3.1 Pro	Google	31.9
6	Grok 4.6	xAI	30.5
7	Gemini 3.8 Flash	Google	29.6
8	Claude Opus 4.8	Anthropic	28.8
9	Muse Spark 1.1	Meta	28.1
10	Claude Opus 4.7 (Adaptive)	Anthropic	27.3
11	Muse Spark 1.2	Meta	27.2
12	Gemini 3.7 Flash	Google	26.5
13	Grok 4.5	xAI	25.3
14	Muse Spark 1.3	Meta	25.0
15	Gemini 3.6 Flash	Google	22.1
16	GPT-5.6 Sol	OpenAI	22.0
17	Gemini 3.5 Flash	Google	21.2
18	GPT-5.5	OpenAI	20.5
19	Kimi K3	Moonshot AI	19.7
20	Grok 4.3	xAI	18.0
21	Claude Sonnet 5	Anthropic	16.5
22	Gemini 3 Pro	Google	15.3
23	Claude Opus 4.7	Anthropic	14.8
24	GLM-5.3	Z.AI	14.3
25	Claude Opus 4.5 Thinking	Anthropic	14.0
26	Claude Opus 4.6 (Adaptive)	Anthropic	13.7
27	Qwen3.7 Max	Alibaba	13.5
28	Qwen3.8 Max Preview	Alibaba	12.0
29	GPT-5.3 Codex	OpenAI	10.9
30	Qwen 3.6 Max (preview)	Alibaba	9.2
31	GLM-5.3-Flash	Z.AI	7.5
32	Muse Spark	Meta	7.2
33	GPT-5.4	OpenAI	5.8
34	GPT-5.1	OpenAI	5.4
35	Kimi K2.6	Moonshot AI	5.3
36	Gemini 3.5 Flash-Lite	Google	5.2
37	MiMo-V2-Pro	Xiaomi	4.6
38	GLM-5.2	Z.AI	4.4
39	MiMo-V2.5-Pro	Xiaomi	3.3
40	Claude Opus 4.6	Anthropic	2.4
41	Grok 4	xAI	2.1
42	TMInkling	Thinking Machines Lab	2.0
43	MiniMax M3	MiniMax	1.4
44	Qwen3.7 Plus	Alibaba	1.1
45	GLM-5.1	Z.AI	0.9
45	Qwen3.6 Plus	Alibaba	0.9
47	DeepSeek V4 Pro 0813	DeepSeek	0.8
47	MiniMax M2.7	MiniMax	0.8
49	GLM-5	Z.AI	0.3
50	GPT-5.6 Terra	OpenAI	0.1
51	Nemotron 3 Ultra	NVIDIA	-0.4
52	Solar Pro 4	Upstage	-0.8
53	GPT-5.2	OpenAI	-0.9
54	GPT-5.2-Codex	OpenAI	-2.2
55	MCQuasar 438B	Multiverse Computing	-2.6
56	Claude Sonnet 4.6	Anthropic	-3.5
57	Command A+	Cohere	-4.0
58	Claude Opus 4.5	Anthropic	-4.1
59	Gemini 3 Flash	Google	-4.3
60	Ling 3.0 Flash VL	InclusionAI	-4.5
61	DeepSeek V4.1 Flash	DeepSeek	-5.3
62	GPT-5.1-Codex	OpenAI	-6.5
62	GPT-5.1-Codex-Max	OpenAI	-6.5
64	K-EXAONE 2.0	LG AI Research	-6.6
65	Kimi K2.5	Moonshot AI	-7.3
65	Kimi K2.5 (Reasoning)	Moonshot AI	-7.3
67	STA.X K2	SK Telecom	-8.2
68	GPT-5 (high)	OpenAI	-8.7
69	TMInkling-Small	Thinking Machines Lab	-8.9
70	Claude 4 Sonnet	Anthropic	-9.0
71	Qwen3.8-Flash-Next	Alibaba	-9.7
72	Qwen3.8-27B	Alibaba	-10.0
73	Kimi K2.7 Code	Moonshot AI	-10.2
74	GPT-5.6 Luna	OpenAI	-10.3
75	GPT-4o	OpenAI	-10.5
76	GPT-5 (medium)	OpenAI	-10.9
77	o1	OpenAI	-11.0
78	OPMiniCPM5-2B	OpenBMB	-11.6
79	Granite 4.2 30B	IBM	-12.9
80	DeepSeek V4 Flash 0731	DeepSeek	-14.3
81	Granite 4.2 3B	IBM	-14.7
82	o3	OpenAI	-15.5
83	Gemini 2.5 Pro	Google	-16.3
84	GLM-5-Turbo	Z.AI	-16.4
85	Llama 3.1 405B	Meta	-17.1
86	Granite 4.2 8B	IBM	-17.2
87	Nemotron 3.5 Lightning 30B A3B NVFP4	NVIDIA	-17.7
88	Ling 3.0 Flash	InclusionAI	-17.9
88	Ling 3.0 Flash FP8	InclusionAI	-17.9
90	Hy3	Tencent	-18.5
90	Hy3 Preview	Tencent	-18.5
92	GPT-5.4 mini	OpenAI	-18.9
93	GLM-5V-Turbo	Z.AI	-19.3
93	Ling 3.0 Tiny	InclusionAI	-19.3
95	Gemma 4 E4B	Google	-19.7
96	Qwen3.6-27B	Alibaba	-20.0
97	MiMo-V2-Omni	Xiaomi	-20.1
98	APApodex 1.1	Apodex	-21.9
98	APApodex 1.1 Mini	Apodex	-21.9
100	Qwen3.6-35B-A3B	Alibaba	-22.2
101	Gemma 4 E2B	Google	-23.6
102	DeepSeek-R1	DeepSeek	-27.4
103	Kimi K2	Moonshot AI	-28.3
104	GPT-5.4 nano	OpenAI	-29.5
104	LFM2.5-2.6B	LiquidAI	-29.5
106	DeepSeek V3.1 (Reasoning)	DeepSeek	-29.6
107	Grok 4 Fast (Reasoning)	xAI	-29.9
107	Grok 4.1 Fast (Reasoning)	xAI	-29.9
109	Mistral Small 4	Mistral	-30.4
109	Mistral Small 4 (Reasoning)	Mistral	-30.4
111	Mistral Medium 3	Mistral	-31.4
112	GLM-4.6	Z.AI	-31.7
113	Muse Glimmer 30B	Meta	-32.8
114	LFM2.5-8B-A1B	LiquidAI	-33.3
115	Mistral Large 2	Mistral	-34.4
116	GLM-4.7	Z.AI	-36.4
117	Mistral Medium 3.5 128B	Mistral	-36.8
118	Grok Code Fast 1	xAI	-37.1
119	Step 3.7 Flash	StepFun	-37.3
120	Qwen3.5 397B	Alibaba	-37.9
120	Qwen3.5 397B (Reasoning)	Alibaba	-37.9
122	GPT-4.1	OpenAI	-39.6
122	Mistral Large 3	Mistral	-39.6
124	Nemotron 3 Super 100B	NVIDIA	-41.5
124	Qwen3.5-122B-A10B	Alibaba	-41.5
126	DeepSeek V3	DeepSeek	-41.6
127	Llama 4 Maverick	Meta	-41.8
128	Gemini 2.5 Flash	Google	-42.6
129	DeepSeek V3.1	DeepSeek	-42.7
130	Qwen3 Max	Alibaba	-43.5
131	Qwen3.5-27B	Alibaba	-44.0
132	Trinity-Large-Preview	Arcee AI	-44.1
132	Trinity-Large-Thinking	Arcee AI	-44.1
134	Nemotron Ultra 253B	NVIDIA	-44.9
135	DeepSeek V3.2	DeepSeek	-46.9
136	Nova Pro	Amazon	-47.7
137	Gemma 4 31B	Google	-47.9
138	Qwen3.5-35B-A3B	Alibaba	-48.1
139	MiMo-V2-Flash	Xiaomi	-48.4
140	Claude 3 Haiku	Anthropic	-48.6
140	North Mini Code	Cohere	-48.6
142	GPT-OSS 120B	OpenAI	-49.2
143	Gemma 4 26B A4B	Google	-50.8
144	Grok 4.1 Fast	xAI	-50.9
145	Nemotron 3 Nano 30B	NVIDIA	-51.6
146	Llama 4 Scout	Meta	-52.1
147	Gemma 4 12B	Google	-52.7
148	Solar Pro 3	Upstage	-53.4
149	GPT-4.1 mini	OpenAI	-53.6
150	FAUltravox v0.6 Llama 3.3 70B	Fixie AI	-54.2
151	Phi-4	Microsoft	-55.7
152	Nemotron 3 Nano Omni 30B A3B	NVIDIA	-57.4
153	GPT-4.1 nano	OpenAI	-57.6
154	K-Exaone	LG AI Research	-58.0
155	SASarvam 105B	Sarvam	-59.4
156	Qwen3-Omni-30B-A3B-Thinking	Alibaba	-61.4
157	GLM-4.5-Air	Z.AI	-61.5
158	Solar Pro 2	Upstage	-62.0
159	Exaone 4.0 32B	LG AI Research	-62.8
160	GPT-OSS 20B	OpenAI	-63.0
161	Ling 2.6 Flash	InclusionAI	-66.1
162	Gemma 3 27B	Google	-67.2
163	Granite-4.0-350M	IBM	-69.3
163	Qwen3-Omni-30B-A3B-Instruct	Alibaba	-69.3
165	SASarvam 30B	Sarvam	-71.5
166	CECeleris-1	Celeris	-71.6
167	Granite-4.0-H-1B	IBM	-72.3
168	Granite-4.0-H-350M	IBM	-80.9
169	Granite-4.0-1B	IBM	-81.6
170	Exaone 4.0 1.2B	LG AI Research	-82.1
171	LFM2.5-VL-1.6B-Extract	LiquidAI	-84.4

Evidence key: ObservedLast good

Rows are ordered by the value AA-Omniscience: Knowledge and Hallucination Benchmark published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.