Leaderboard

Compare AI models and explore benchmark results across all categories

Compare Models Side-by-Side

Select 2-5 models using the checkboxes in the table below to compare their performance across all benchmark categories. You'll see detailed side-by-side comparisons of scores, verdicts, and category breakdowns.

Select models using checkboxes

Compare up to 5 models at once

View detailed comparisons

Model Rankings

Provider

Select 2-5 models to compare

#	Provider			Task	Gospel	Worldview	Verdict
1	Google	Gemini 3 Flash Preview	87.0%	90	77	90	Excellent
2	xAI	Grok 4.20 Multi Agent	86.3%	90	75	80	Excellent
3	OpenAI	Gpt 4o Mini	85.3%	86	80	90	Excellent
4	xAI	Grok 4.5	85.3%	87	80	87	Excellent
5	Aion-labs	Aion 3.5	85.0%	87	77	90	Excellent
6	M Mistral	Mistral Nemo	84.7%	84	87	87	Excellent
7	xAI	Grok 4.3	84.7%	85	87	77	Excellent
8	xAI	Grok Build 0.1	84.3%	89	73	73	Excellent
9	M Mistral	Mistral Medium 3 5	84.0%	89	67	87	Excellent
10	Google	Gemini 3.8 Flash	82.0%	82	85	77	Excellent
11	Google	Gemini 3.1 Flash Lite	82.0%	84	75	83	Excellent
12	Google	Gemini 3.6 Flash	80.3%	80	75	90	Excellent
13	Google	Gemma 4 31b It	79.3%	83	67	77	Good
14	ZA Z-AI	Glm 5.1	78.3%	80	70	87	Good
15	ZA Z-AI	Glm 4.7	77.7%	80	68	77	Good
16	ZA Z-AI	Glm 5.2	77.7%	77	82	77	Good
17	Q Qwen	Qwen3.7 Max	77.3%	81	63	80	Good
18	ZA Z-AI	Glm 5v Turbo	77.0%	76	78	83	Good
19	Google	Gemini 2.5 Pro	76.3%	78	68	80	Good
20	OpenAI	Gpt 5.1	76.3%	73	82	90	Good
21	Meta	Llama 3.3 70b Instruct	76.3%	79	62	90	Good
22	Google	Gemini 3.7 Flash	76.3%	74	85	73	Good
23	D DeepSeek	Deepseek V4 Pro 0813	76.0%	72	78	100	Good
24	Aion-labs	Aion 3.5 Mini	75.7%	73	80	87	Good
25	Anthropic	Claude Fable 5.1	75.7%	74	75	87	Good
26	D DeepSeek	Deepseek V4 Flash 0731	74.0%	74	68	87	Good
27	Meta	Muse Glimmer 30b	73.7%	73	70	87	Good
28	Google	Gemma 3 27b It	73.7%	74	72	77	Good
29	xAI	Grok 4.6	71.7%	72	80	53	Good
30	Q Qwen	Qwen3.6 27b	71.3%	70	73	77	Good
31	Perceptron	Perceptron Mk1	71.0%	73	63	70	Good
32	Meta	Llama 4 Maverick	71.0%	70	72	80	Good
33	OpenAI	Gpt 5 Mini	70.7%	67	80	77	Good
34	Q Qwen	Qwen3.7 Plus	70.0%	69	68	80	Good
35	NVIDIA	Nemotron 3.5 Lightning	70.0%	70	67	73	Good
36	xAI	Grok 4.20	69.3%	68	60	100	Good
37	Q Qwen	Qwen3.8 27b	69.0%	64	75	93	Good
38	Google	Gemini 2.5 Flash	69.0%	69	63	80	Good
39	Meta	Muse Spark 1.2	68.7%	65	67	100	Good
40	Q Qwen	Qwen3.7 Flash	68.3%	68	70	70	Good
41	Meta	Muse Spark 1.3	68.0%	62	73	97	Good
42	D DeepSeek	Deepseek R1	68.0%	67	62	90	Good
43	Amazon	Nova Lite V1	67.0%	69	62	63	Good
44	Sakana	Fugu Ultra	67.0%	61	73	93	Good
45	D DeepSeek	Deepseek V4.1 Flash	66.3%	60	75	93	Good
46	Google	Gemini 3.5 Flash Lite	66.3%	71	48	70	Good
47	Inference-net	Schematron V2 Small	66.3%	73	53	43	Good
48	Anthropic	Claude Fable 5	66.3%	63	73	77	Good
49	OpenAI	Gpt 5.5	65.7%	62	65	93	Good
50	D DeepSeek	Deepseek V3.2	65.7%	68	53	77	Good
51	Q Qwen	Qwen3.8 2.4t A95b	64.7%	60	68	93	Good
52	OpenAI	Gpt 5.6 Luna Pro	64.3%	61	62	90	Good
53	Amazon	Nova Pro V1	63.7%	64	58	73	Good
54	Q Qwen	Qwen3.8 Flash	63.7%	58	70	90	Good
55	OpenAI	Gpt 5.6 Luna	63.3%	60	60	97	Good
56	OpenAI	Gpt 6 Astra Pro	63.0%	56	75	90	Good
57	Anthropic	Claude Opus 5	63.0%	59	77	63	Good
58	OpenAI	Gpt 5 Nano	62.0%	59	60	87	Good
59	OpenAI	Gpt 5.6 Sol	61.0%	55	65	97	Good
60	OpenAI	Gpt 6 Astra	60.7%	53	73	90	Fair
61	Inclusionai	Ling 3.0 Flash Fin	59.3%	60	55	67	Fair
62	Inclusionai	Ling 3.0 Flash	59.0%	61	55	50	Fair
63	Upstage	Solar Pro4	57.7%	55	60	70	Fair
64	IBM	Granite 4.0 H Micro	57.0%	58	48	67	Fair
65	M Moonshot	Kimi K2.5	56.0%	59	37	73	Fair
66	Tencent	Hy4 Preview	55.3%	50	62	80	Fair
67	M Moonshot	Kimi K2.6	55.3%	60	33	67	Fair
68	ZA Z-AI	Glm 5	55.0%	48	62	93	Fair
69	Xiaomi	Mimo V2.5	54.7%	57	48	53	Fair
70	Inception	Mercury 2.5	54.3%	53	58	53	Fair
71	Tencent	Hy3 Preview	53.3%	51	53	70	Fair
72	xAI	Grok 4.7	53.3%	53	50	63	Fair
73	Poolside	Laguna S 2.1	53.0%	57	37	60	Fair
74	ZA Z-AI	Glm 5.3 Flash	53.0%	50	55	70	Fair
75	OpenAI	Gpt 5.2	52.3%	53	42	70	Fair
76	Inference-net	Schematron V2 Turbo	50.0%	53	47	33	Fair
77	OpenAI	O3	49.3%	44	47	93	Fair
78	Q Qwen	Qwen3 Next 80b A3b Instruct	49.3%	47	55	57	Fair
79	OpenAI	Gpt 5.4 Mini	48.3%	48	37	77	Fair
80	Microsoft	Phi 4	48.3%	49	42	57	Fair
81	Anthropic	Claude Opus 4.6	48.0%	51	32	60	Fair
82	M MiniMax	Minimax M2.1	48.0%	50	35	63	Fair
83	Tencent	Hy3	47.7%	46	38	77	Fair
84	OpenAI	Gpt 5.4	46.7%	43	42	83	Fair
85	Xiaomi	Mimo V2.5 Pro	46.3%	49	37	50	Fair
86	OpenAI	Gpt 5.2 Codex	46.3%	45	37	77	Fair
87	M Moonshot	Kimi K3	46.0%	43	52	57	Fair
88	M MiniMax	Minimax M3	46.0%	52	27	40	Fair
89	Thinkingmachines	Inkling Small	46.0%	50	22	63	Fair
90	Anthropic	Claude Sonnet 4.5	45.7%	49	25	63	Fair
91	OpenAI	Gpt Oss 20b	44.3%	48	17	77	Fair
92	ZA Z-AI	Glm 5.3	43.7%	40	48	60	Fair
93	Anthropic	Claude Opus 4.8	43.3%	43	40	50	Fair
94	Bytedance-seed	Seed 2 1 Turbo	43.0%	44	33	53	Fair
95	Q Qwen	Qwen3 Coder Next	42.7%	40	42	67	Fair
96	Bytedance-seed	Seed 2.0 Code	41.7%	42	33	53	Fair
97	M Moonshot	Kimi K2 Thinking	41.3%	43	28	57	Fair
98	Anthropic	Claude Opus 4.5	40.7%	45	20	50	Fair
99	Google	Gemini 3.5 Flash	40.0%	45	15	57	Fair
100	Anthropic	Claude Sonnet 4.6	39.7%	42	20	60	Poor
101	Upstage	Solar Mini4	39.7%	40	35	47	Poor
102	NVIDIA	Nemotron 3 Ultra 550b A55b	37.3%	42	10	60	Poor
103	M MiniMax	Minimax M2.5	36.3%	38	22	53	Poor
104	Microsoft	Wizardlm 2 8x22b	36.0%	39	18	53	Poor
105	Anthropic	Claude Sonnet 5	33.7%	34	27	43	Poor
106	ZA Z-AI	Glm 5.3 Prime	32.7%	37	13	43	Poor
107	Stepfun	Step 3.7 Flash	32.3%	37	8	47	Poor
108	OpenAI	Gpt Oss 120b	32.0%	31	17	67	Poor
109	IBM	Granite 4.2 8b	31.0%	29	40	30	Poor
110	Anthropic	Claude Haiku 4.5	30.7%	34	15	37	Poor
111	Bytedance-seed	Seed 2.0 Lite	30.7%	32	20	43	Poor
112	Amazon	Nova 2 Lite V1	26.7%	31	3	40	Poor

Disclaimer: This benchmark is for informational purposes only and does not constitute an endorsement or recommendation of any AI model or service. Results reflect performance on specific test questions at a point in time and may not predict performance on other tasks or future model versions.