Composite Benchmark Builder

Live methodology

1 Source Benchmark

Snapshot 7/31/2026, 11:33:21 PM

Intelligence Index × 10–100 Composite Score

Result · updates immediately

Leaderboard

614 ranked · 331 excluded

614 ranked models and 331 excluded models. Select a model name to inspect its score evidence.
	Change								Context	Decision
1	—		Anthropic	100.0	100%	0%	10	56.2	1M	pareto
2	—	xhigh reasoning	Anthropic	99.8	100%	0%	10	54.7	—	dominated
3	—		Anthropic	99.7	100%	0%	20	62.7	1M	dominated
4	—	high reasoning	Anthropic	99.4	100%	0%	10	52.3	—	dominated
4	—	max reasoning	OpenAI	99.4	100%	0%	11.3	69.3	1.1M	dominated
6	—	xhigh reasoning	OpenAI	99.2	100%	0%	11.3	62.8	—	dominated
7	—	max reasoning	Kimi	99.0	100%	0%	6	33.1	—	pareto
8	—	medium reasoning	Anthropic	98.9	100%	0%	10	55.2	—	dominated
9	—	high reasoning	OpenAI	98.7	100%	0%	11.3	59.2	—	dominated
10	—		Anthropic	98.5	100%	0%	10	—	1M	dominated
11	—	max reasoning	OpenAI	98.4	100%	0%	4.5	139.1	1.1M	pareto
12	—	xhigh reasoning	OpenAI	98.2	100%	0%	11.3	—	1.1M	dominated
13	—	high reasoning	SpaceXAI	98.0	100%	0%	3	55	500K	pareto
14	—	medium reasoning	OpenAI	97.8	100%	0%	11.3	56.3	—	dominated
14	—		OpenAI	97.8	100%	0%	4.8	90.5	400K	dominated
16	—		Anthropic	97.6	100%	0%	10	—	—	dominated
17	—		Anthropic	97.4	100%	0%	4	79.2	1M	dominated
18	—	high reasoning	OpenAI	97.2	100%	0%	11.3	—	—	dominated
19	—		Alibaba	97.1	100%	0%	2.9	33.7	262.1K	pareto
20	—	xhigh reasoning	OpenAI	96.9	100%	0%	4.5	109.7	—	dominated
21	—	xhigh reasoning	OpenAI	96.7	100%	0%	5.6	—	1.1M	dominated
22	—	max reasoning	OpenAI	96.6	100%	0%	0.5	174.2	1.1M	pareto
23	—	max reasoning	Z AI	96.4	100%	0%	2.2	107.3	—	dominated
24	—	low reasoning	Anthropic	96.2	100%	0%	10	51.9	—	dominated
24	—	xhigh reasoning	Meta	96.2	100%	0%	2	146.4	1.0M	dominated
26	—	medium reasoning	OpenAI	95.9	100%	0%	11.3	—	—	dominated
27	—	high reasoning	Google	95.8	100%	0%	3.4	220.6	1.0M	dominated
28	—	high reasoning	Google	95.6	100%	0%	3	220.4	1.0M	dominated
29	—		DeepSeek	95.4	100%	0%	0.2	—	—	pareto
30	—		MiniMax	95.3	100%	0%	0.5	46	204.8K	dominated
31	—	low reasoning	OpenAI	95.1	100%	0%	11.3	57.7	—	dominated
32	—	xhigh reasoning	OpenAI	94.9	100%	0%	0.5	174	—	dominated
33	—	high reasoning	OpenAI	94.7	100%	0%	4.5	116.3	—	dominated
33	—		Xiaomi	94.7	100%	0%	0.8	—	1.0M	dominated
35	—		OpenAI	94.5	100%	0%	1.7	174.4	400K	dominated
36	—		Anthropic	94.3	100%	0%	6	—	—	dominated
37	—		Z AI	94.1	100%	0%	1.5	—	202.8K	dominated
38	—	low reasoning	Kimi	94.0	100%	0%	6	33.7	—	dominated
39	—		Google	93.8	100%	0%	4.5	135.6	1.0M	dominated
40	—	high reasoning	OpenAI	93.6	100%	0%	0.5	172.1	—	dominated
41	—	max reasoning	Alibaba	93.5	100%	0%	3.8	200.2	1M	dominated
42	—	medium reasoning	OpenAI	93.3	100%	0%	4.5	98.4	—	dominated
43	—	medium reasoning	Google	93.1	100%	0%	3.4	240.9	—	dominated
44	—		Alibaba	93.0	100%	0%	1.3	50.9	262.1K	dominated
45	—		MiniMax	92.8	100%	0%	0.5	74.1	524.3K	dominated
46	—		DeepSeek	92.6	100%	0%	0.5	61.4	1.0M	dominated
46	—	xhigh reasoning	OpenAI	92.6	100%	0%	4.8	109.1	—	dominated
48	—		Kimi	92.3	100%	0%	1.7	—	256K	dominated
49	—		Motif Technologies	92.2	100%	0%	0	—	—	pareto
50	—		OpenAI	92.0	100%	0%	0.5	163.2	400K	dominated
51	—		KwaiKAT	91.8	100%	0%	0.5	111.5	256K	dominated
52	—		Anthropic	91.7	100%	0%	10	—	—	dominated
53	—	low reasoning	OpenAI	91.5	100%	0%	11.3	—	—	dominated
54	—		Xiaomi	91.4	100%	0%	0.8	—	262.1K	dominated
55	—	high reasoning	DeepSeek	91.1	100%	0%	0.5	55.3	—	dominated
55	—		Meta	91.1	100%	0%	0	—	—	dominated
57	—		Z AI	90.9	100%	0%	1.9	—	202.8K	dominated
58	—	non-reasoning reasoning	Anthropic	90.7	100%	0%	10	—	—	dominated
59	—	xhigh reasoning	OpenAI	90.5	100%	0%	4.8	—	400K	dominated
59	—		Xiaomi	90.5	100%	0%	0.5	46.5	1.0M	dominated
61	—		Kimi	90.1	100%	0%	1.7	39.6	—	dominated
61	—		Tencent	90.1	100%	0%	0	86.4	262.1K	dominated
63	—	non-reasoning reasoning	Anthropic	89.9	100%	0%	4	65	—	dominated
64	—	non-reasoning reasoning	OpenAI	89.6	100%	0%	11.3	60.2	—	dominated
64	—		Tencent	89.6	100%	0%	0.2	72.2	—	dominated
66	—		Nex AGI	89.4	100%	0%	1	130.7	—	dominated
67	—		Anthropic	89.2	100%	0%	10	—	—	dominated
68	—	xhigh reasoning	Thinking Machines	89.1	100%	0%	2.6	79.6	1.0M	dominated
69	—	low reasoning	OpenAI	88.9	100%	0%	4.5	109.6	—	dominated
70	—		DeepSeek	88.7	100%	0%	0.2	106.7	1.0M	dominated
70	—		Xiaomi	88.7	100%	0%	0	—	1.0M	dominated
72	—		Z AI	88.3	100%	0%	2.1	—	202.8K	dominated
72	—		Thinking Machines	88.3	100%	0%	0.5	95.2	524.3K	dominated
74	—	xhigh reasoning	OpenAI	88.1	100%	0%	4.8	—	400K	dominated
75	—	xhigh reasoning	OpenAI	87.8	100%	0%	1.7	—	400K	dominated
75	—	max reasoning	Alibaba	87.8	100%	0%	2.9	—	—	dominated
77	—		SpaceXAI	87.6	100%	0%	1.3	—	—	dominated
78	—	high reasoning	Google	87.4	100%	0%	4.5	—	—	dominated
78	—		Alibaba	87.4	100%	0%	1.1	55	1M	dominated
80	—		Z AI	87.1	100%	0%	1.6	—	204.8K	dominated
81	—	low reasoning	OpenAI	86.9	100%	0%	5.6	—	—	dominated
82	—		Alibaba	86.8	100%	0%	0.7	51.7	1M	dominated
83	—		Sapiens AI	86.5	100%	0%	0.6	127.2	—	dominated
83	—		China Mobile	86.5	100%	0%	0	—	—	dominated
85	—	xhigh reasoning	OpenAI	86.3	100%	0%	0.5	—	400K	dominated
86	—		Z AI	86.0	100%	0%	0	—	—	dominated
86	—	medium reasoning	OpenAI	86.0	100%	0%	0.5	158.4	—	dominated
86	—		MiniMax	86.0	100%	0%	0.5	—	—	dominated
89	—	medium reasoning	OpenAI	85.6	100%	0%	4.8	—	—	dominated
90	—		Anthropic	85.3	100%	0%	10	—	1M	dominated
90	—		Google	85.3	100%	0%	1.1	—	—	dominated
90	—		NVIDIA	85.3	100%	0%	1.2	222.3	1M	dominated
93	—	high reasoning	SpaceXAI	85.0	100%	0%	1.6	—	1M	dominated
94	—	high reasoning	DeepSeek	84.8	100%	0%	0.2	—	—	dominated
95	—		Xiaomi	84.7	100%	0%	0.2	76.4	—	dominated
96	—		Anthropic	84.4	100%	0%	6	49.2	1M	dominated
96	—		Alibaba	84.4	100%	0%	1.4	58	262.1K	dominated
98	—		SpaceXAI	84.2	100%	0%	3	—	—	dominated
99	—	high reasoning	OpenAI	84.0	100%	0%	3.4	—	400K	dominated
100	—		Google	83.8	100%	0%	0.9	351.2	1.0M	dominated
100	—		SpaceXAI	83.8	100%	0%	3	—	2M	dominated
102	—		Anthropic	83.4	100%	0%	6	—	—	dominated
102	—		Xiaomi	83.4	100%	0%	0	—	—	dominated
104	—		ByteDance Seed	83.2	100%	0%	0	—	—	dominated
105	—	high reasoning	OpenAI	83.0	100%	0%	3.4	—	400K	dominated
106	—		Anthropic	82.8	100%	0%	30	38.1	200K	dominated
106	—	medium reasoning	SpaceXAI	82.8	100%	0%	1.6	129.8	—	dominated
108	—		Anthropic	82.5	100%	0%	6	—	1M	dominated
109	—	non-reasoning reasoning	Z AI	82.1	100%	0%	2.1	—	—	dominated
109	—	non-reasoning reasoning	OpenAI	82.1	100%	0%	11.3	—	—	dominated
109	—	low reasoning	SpaceXAI	82.1	100%	0%	1.6	137.5	—	dominated
109	—		Kimi	82.1	100%	0%	1.2	—	262.1K	dominated
113	—		Google	81.6	100%	0%	1.1	179.3	1.0M	dominated
113	—		Xiaomi	81.6	100%	0%	0	—	—	dominated
115	—	minimal reasoning	Google	81.4	100%	0%	3.4	223.3	—	dominated
116	—	non-reasoning reasoning	Anthropic	81.1	100%	0%	10	—	200K	dominated
116	—	high reasoning	OpenAI	81.1	100%	0%	3.4	—	400K	dominated
116	—	high reasoning	OpenAI	81.1	100%	0%	3.4	—	400K	dominated
119	—	non-reasoning reasoning	Kimi	80.8	100%	0%	1.7	37.8	—	dominated
120	—		Z AI	80.6	100%	0%	0	—	—	dominated
121	—		Anthropic	80.4	100%	0%	6	48.1	—	dominated
122	—	non-reasoning reasoning	Z AI	80.3	100%	0%	2.1	64.3	—	dominated
123	—	non-reasoning reasoning	OpenAI	80.1	100%	0%	4.5	102.9	—	dominated
124	—		Alibaba	79.9	100%	0%	0.8	—	262.1K	dominated
125	—		Anthropic	79.4	100%	0%	30	—	—	dominated
125	—		Z AI	79.4	100%	0%	1	—	202.8K	dominated
125	—	medium reasoning	OpenAI	79.4	100%	0%	3.4	—	—	dominated
125	—		KwaiKAT	79.4	100%	0%	0.5	104.6	—	dominated
125	—		MiniMax	79.4	100%	0%	0.5	—	196.6K	dominated
125	—		Alibaba	79.4	100%	0%	1.4	64.9	—	dominated
131	—		Tencent	78.8	100%	0%	0.1	163.5	262.1K	dominated
132	—		OpenAI	78.5	100%	0%	11.3	—	—	dominated
132	—		LongCat	78.5	100%	0%	1.3	45.5	—	dominated
134	—	low reasoning	OpenAI	78.2	100%	0%	0.5	162.1	—	dominated
134	—		SpaceXAI	78.2	100%	0%	6	—	256K	dominated
136	—		Xiaomi	78.0	100%	0%	0	—	—	dominated
137	—	low reasoning	Google	77.8	100%	0%	4.5	—	—	dominated
138	—		Anthropic	77.6	100%	0%	30	37.9	200K	dominated
138	—		Anthropic	77.6	100%	0%	6	49.4	1M	dominated
140	—		Kimi	77.3	100%	0%	1.1	—	262.1K	dominated
141	—		OpenAI	77.2	100%	0%	35	—	200K	dominated
142	—	non-reasoning reasoning	Z AI	77.0	100%	0%	1.6	—	—	dominated
143	—		Alibaba	76.8	100%	0%	1.1	137.9	262.1K	dominated
144	—		DeepSeek	76.6	100%	0%	0.3	—	—	dominated
144	—	non-reasoning reasoning	Alibaba	76.6	100%	0%	1.4	64.3	—	dominated
146	—		Alibaba	76.3	100%	0%	1.6	—	262.1K	dominated
147	—		Alibaba	76.2	100%	0%	0.6	131.1	262.1K	dominated
148	—		MiniMax	76.0	100%	0%	0.5	—	196.6K	dominated
149	—	non-reasoning reasoning	DeepSeek	75.7	100%	0%	0.5	63.8	—	dominated
149	—	low reasoning	OpenAI	75.7	100%	0%	3.4	—	—	dominated
149	—		Xiaomi	75.7	100%	0%	0.2	—	—	dominated
152	—		Anthropic	75.4	100%	0%	2	109.9	200K	dominated
153	—		Anthropic	75.2	100%	0%	30	—	—	dominated
154	—	medium reasoning	OpenAI	75.0	100%	0%	0.7	—	—	dominated
155	—	high reasoning	OpenAI	74.6	100%	0%	0.7	—	400K	dominated
155	—		SpaceXAI	74.6	100%	0%	0	—	—	dominated
155	—		Alibaba	74.6	100%	0%	1.5	53	—	dominated
155	—		InclusionAI	74.6	100%	0%	0.9	125.7	262.1K	dominated
159	—	non-reasoning reasoning	Alibaba	74.2	100%	0%	1.4	57.6	—	dominated
160	—		DeepSeek	74.0	100%	0%	1.9	—	—	dominated
160	—		OpenAI	74.0	100%	0%	3.5	142.3	200K	dominated
162	—		StepFun	73.7	100%	0%	0.4	394.3	—	dominated
163	—	medium reasoning	OpenAI	73.6	100%	0%	0.5	—	—	dominated
164	—		Mistral	73.4	100%	0%	3	72.9	262.1K	dominated
165	—	medium reasoning	OpenAI	73.2	100%	0%	1.7	—	—	dominated
166	—		Anthropic	73.1	100%	0%	2	140.1	—	dominated
167	—		Google	72.8	100%	0%	0	35.4	—	dominated
167	—	non-reasoning reasoning	Kimi	72.8	100%	0%	1.2	—	—	dominated
169	—	non-reasoning reasoning	Anthropic	72.4	100%	0%	6	—	—	dominated
169	—	non-reasoning reasoning	Alibaba	72.4	100%	0%	0.8	—	—	dominated
169	—		Alibaba	72.4	100%	0%	0.7	—	262.1K	dominated
172	—		Anthropic	72.0	100%	0%	6	—	—	dominated
172	—		OpenAI	72.0	100%	0%	11.3	—	—	dominated
174	—	non-reasoning reasoning	DeepSeek	71.7	100%	0%	0.2	106.4	—	dominated
174	—		Z AI	71.7	100%	0%	1	—	—	dominated
176	—		China Mobile	71.5	100%	0%	0	—	—	dominated
177	—		KwaiKAT	71.2	100%	0%	0	—	—	dominated
177	—		MiniMax	71.2	100%	0%	0.5	—	196.6K	dominated
179	—	non-reasoning reasoning	Anthropic	71.0	100%	0%	30	—	—	dominated
180	—	non-reasoning reasoning	Xiaomi	70.8	100%	0%	0.5	46.3	—	dominated
181	—	non-reasoning reasoning	OpenAI	70.6	100%	0%	5.6	—	—	dominated
182	—	non-reasoning reasoning	Alibaba	70.5	100%	0%	1.1	153.2	—	dominated
183	—	non-reasoning reasoning	Google	70.2	100%	0%	1.1	—	—	dominated
183	—		SpaceXAI	70.2	100%	0%	0.3	—	—	dominated
185	—		Anthropic	70.0	100%	0%	0	—	—	dominated
186	—	non-reasoning reasoning	Z AI	69.7	100%	0%	1	—	—	dominated
186	—	non-reasoning reasoning	OpenAI	69.7	100%	0%	0.5	165.6	—	dominated
188	—		Z AI	69.5	100%	0%	0.6	48.9	131.1K	dominated
189	—	non-reasoning reasoning	Tencent	69.2	100%	0%	0.1	163.6	—	dominated
189	—		InclusionAI	69.2	100%	0%	0.9	—	262.1K	dominated
191	—		ByteDance Seed	68.8	100%	0%	0	—	—	dominated
191	—	non-reasoning reasoning	OpenAI	68.8	100%	0%	4.8	—	—	dominated
191	—		StepFun	68.8	100%	0%	0.2	—	—	dominated
194	—		Google	68.5	100%	0%	3.4	131.3	1.0M	dominated
195	—		Google	68.4	100%	0%	0.2	—	262.1K	dominated
196	—	high reasoning	OpenAI	68.2	100%	0%	1.9	—	200K	dominated
197	—	non-reasoning reasoning	Anthropic	67.9	100%	0%	30	—	—	dominated
197	—	non-reasoning reasoning	Anthropic	67.9	100%	0%	6	—	—	dominated
197	—		StepFun	67.9	100%	0%	0.2	—	256K	dominated
200	—		DeepSeek	67.5	100%	0%	0.3	—	—	dominated
200	—		NVIDIA	67.5	100%	0%	0.4	146.6	262.1K	dominated
202	—	high reasoning	OpenAI	67.2	100%	0%	0.7	—	400K	dominated
203	—		Google	67.0	100%	0%	0.6	302.4	1.0M	dominated
203	—		Alibaba	67.0	100%	0%	2.4	—	—	dominated
205	—	non-reasoning reasoning	SpaceXAI	66.7	100%	0%	1.6	119.8	—	dominated
206	—	non-reasoning reasoning	DeepSeek	66.5	100%	0%	0.3	—	163.8K	dominated
206	—	non-reasoning reasoning	Xiaomi	66.5	100%	0%	0	—	262.1K	dominated
208	—	non-reasoning reasoning	Alibaba	66.2	100%	0%	0.8	101.2	—	dominated
209	—	non-reasoning reasoning	Alibaba	66.0	100%	0%	0.7	175.9	—	dominated
209	—	max reasoning	Alibaba	66.0	100%	0%	2.4	—	262.1K	dominated
211	—		Google	65.7	100%	0%	0	—	—	dominated
211	—	high reasoning	OpenAI	65.7	100%	0%	0.3	193.6	131.1K	dominated
213	—	non-reasoning reasoning	Anthropic	65.4	100%	0%	2	103.8	—	dominated
214	—	non-reasoning reasoning	Anthropic	65.2	100%	0%	6	—	200K	dominated
214	—		Kimi	65.2	100%	0%	1.1	—	131.1K	dominated
216	—		OpenAI	64.9	100%	0%	26.3	—	200K	dominated
217	—		Google	64.7	100%	0%	0	—	—	dominated
217	—	non-reasoning reasoning	Z AI	64.7	100%	0%	1	—	202.8K	dominated
219	—		Z AI	64.4	100%	0%	0.2	—	202.8K	dominated
220	—		Cohere	64.1	100%	0%	0	195.6	—	dominated
220	—	high reasoning	SpaceXAI	64.1	100%	0%	0.4	—	—	dominated
220	—	non-reasoning reasoning	SpaceXAI	64.1	100%	0%	3	—	—	dominated
223	—		Google	63.8	100%	0%	3.4	—	—	dominated
224	—		DeepSeek	63.6	100%	0%	0	—	163.8K	dominated
225	—		LG AI Research	63.5	100%	0%	0	—	—	dominated
226	—		Baidu	63.3	100%	0%	0	—	—	dominated
227	—		Google	62.9	100%	0%	0.2	29.5	—	dominated
227	—	non-reasoning reasoning	Google	62.9	100%	0%	0.2	75.2	—	dominated
227	—	non-reasoning reasoning	SpaceXAI	62.9	100%	0%	3	—	—	dominated
227	—	medium reasoning	Amazon	62.9	100%	0%	3.4	138.1	—	dominated
231	—		SpaceXAI	62.5	100%	0%	0	—	256K	dominated
232	—	non-reasoning reasoning	DeepSeek	62.2	100%	0%	0.5	—	163.8K	dominated
232	—		Inception	62.2	100%	0%	0.4	904.6	128K	dominated
232	—		Alibaba	62.2	100%	0%	0.2	66.9	256K	dominated
235	—	non-reasoning reasoning	DeepSeek	61.8	100%	0%	0.3	—	—	dominated
236	—		ServiceNow	61.7	100%	0%	0	—	—	dominated
237	—		Alibaba	61.5	100%	0%	0.6	126.8	262.1K	dominated
238	—	non-reasoning reasoning	DeepSeek	61.3	100%	0%	0.8	—	—	dominated
239	—	medium reasoning	Amazon	61.2	100%	0%	0.9	—	—	dominated
240	—		DeepSeek	61.0	100%	0%	0.9	—	—	dominated
241	—		Alibaba	60.8	100%	0%	2.6	—	—	dominated
242	—		ServiceNow	60.7	100%	0%	0	—	—	dominated
243	—	non-reasoning reasoning	OpenAI	60.5	100%	0%	3.4	—	—	dominated
244	—	non-reasoning reasoning	Alibaba	60.4	100%	0%	0	—	—	dominated
245	—		LG AI Research	60.2	100%	0%	0	—	—	dominated
246	—		DeepSeek	59.8	100%	0%	2.1	—	64K	dominated
246	—		Google	59.8	100%	0%	0.9	—	—	dominated
246	—	non-reasoning reasoning	Google	59.8	100%	0%	0.2	55.9	—	dominated
246	—		Alibaba	59.8	100%	0%	0.1	21.7	—	dominated
250	—	high reasoning	OpenAI	59.4	100%	0%	0.1	—	400K	dominated
251	—		Cohere	59.2	100%	0%	0	21.9	256K	dominated
252	—		Mistral	58.9	100%	0%	0.3	149.9	—	dominated
252	—	low reasoning	Amazon	58.9	100%	0%	3.4	150.4	—	dominated
252	—		Alibaba	58.9	100%	0%	2.6	—	—	dominated
255	—		Z AI	58.6	100%	0%	0	—	—	dominated
256	—		OpenAI	58.3	100%	0%	3.5	—	1.0M	dominated
256	—		Kimi	58.3	100%	0%	1	—	131.1K	dominated
258	—		Mistral	58.0	100%	0%	0	53.2	—	dominated
258	—		Alibaba	58.0	100%	0%	2.4	—	—	dominated
260	—	medium reasoning	OpenAI	57.5	100%	0%	0.1	—	—	dominated
260	—	medium reasoning	Amazon	57.5	100%	0%	0.9	188.4	—	dominated
260	—		OpenAI	57.5	100%	0%	1.9	—	200K	dominated
260	—		Alibaba	57.5	100%	0%	0.3	246.6	—	dominated
264	—		OpenAI	57.1	100%	0%	262.5	—	200K	dominated
265	—	non-reasoning reasoning	Google	56.9	100%	0%	0	—	1.0M	dominated
266	—		DeepSeek	56.7	100%	0%	2.4	—	—	dominated
266	—		China Mobile	56.7	100%	0%	0	—	—	dominated
268	—		SpaceXAI	56.4	100%	0%	8	—	131.1K	dominated
269	—		ByteDance Seed	56.3	100%	0%	0.3	—	—	dominated
270	—	high reasoning	Amazon	56.0	100%	0%	0.9	190.3	—	dominated
270	—		Alibaba	56.0	100%	0%	1.2	—	—	dominated
270	—		Arcee AI	56.0	100%	0%	0.4	209.5	—	dominated
273	—		Alibaba	55.6	100%	0%	3	—	—	dominated
274	—		Mistral	55.4	100%	0%	2.8	45.7	—	dominated
274	—		Alibaba	55.4	100%	0%	2.6	—	—	dominated
276	—		Multiverse Computing	55.0	100%	0%	0.1	418	—	dominated
276	—	low reasoning	Amazon	55.0	100%	0%	0.9	184.1	—	dominated
276	—		Perplexity	55.0	100%	0%	3.5	—	128K	dominated
279	—		MiniMax	54.6	100%	0%	1	—	—	dominated
280	—	non-reasoning reasoning	OpenAI	54.4	100%	0%	0.5	—	—	dominated
280	—		NVIDIA	54.4	100%	0%	0	—	—	dominated
282	—		Google	54.2	100%	0%	0	—	—	dominated
283	—		Mistral	54.0	100%	0%	0	49.4	—	dominated
284	—		MBZUAI Institute of Foundation Models	53.8	100%	0%	0	—	—	dominated
285	—	minimal reasoning	OpenAI	53.6	100%	0%	3.4	—	—	dominated
285	—		LongCat	53.6	100%	0%	0	—	—	dominated
287	—		Naver	53.3	100%	0%	0	—	—	dominated
287	—		OpenAI	53.3	100%	0%	28.9	—	—	dominated
289	—	non-reasoning reasoning	SpaceXAI	53.0	100%	0%	0	—	2M	dominated
290	—		Z AI	52.9	100%	0%	0.5	—	—	dominated
291	—	non-reasoning reasoning	LG AI Research	52.6	100%	0%	0	—	—	dominated
291	—		Alibaba	52.6	100%	0%	1.9	202.9	—	dominated
293	—	non-reasoning reasoning	OpenAI	52.3	100%	0%	1.7	—	—	dominated
293	—	low reasoning	Amazon	52.3	100%	0%	0.9	—	—	dominated
295	—		DeepSeek	51.9	100%	0%	0.5	—	163.8K	dominated
295	—		Z AI	51.9	100%	0%	0.4	—	131.1K	dominated
295	—	non-reasoning reasoning	SpaceXAI	51.9	100%	0%	0.3	—	2M	dominated
298	—		Korea Telecom	51.5	100%	0%	0	—	—	dominated
299	—		InclusionAI	51.4	100%	0%	0	—	—	dominated
300	—		AI9Stars	51.2	100%	0%	0	—	—	dominated
301	—	non-reasoning reasoning	Alibaba	51.1	100%	0%	0.1	17.2	—	dominated
302	—		Mistral	50.9	100%	0%	0.8	63.1	—	dominated
303	—		Prime Intellect	50.7	100%	0%	0	—	131.1K	dominated
303	—	high reasoning	OpenAI	50.7	100%	0%	1.9	—	200K	dominated
305	—	non-reasoning reasoning	Z AI	50.4	100%	0%	0.2	—	—	dominated
306	—		DeepSeek	50.2	100%	0%	0.5	—	—	dominated
307	—		OpenAI	50.1	100%	0%	0	—	—	dominated
308	—		Google	49.8	100%	0%	0.2	—	—	dominated
308	—		SpaceXAI	49.8	100%	0%	0	—	—	dominated
308	—		Upstage	49.8	100%	0%	0	—	—	dominated
311	—		Alibaba	49.4	100%	0%	0.1	71	262.1K	dominated
312	—	low reasoning	OpenAI	49.1	100%	0%	0.3	225	—	dominated
312	—	high reasoning	OpenAI	49.1	100%	0%	0.1	182.5	131.1K	dominated
312	—		NVIDIA	49.1	100%	0%	0.1	319.3	—	dominated
315	—		OpenAI	48.8	100%	0%	0.7	—	1.0M	dominated
316	—		Mistral	48.5	100%	0%	0.8	—	131.1K	dominated
316	—		Mistral	48.5	100%	0%	0.2	—	—	dominated
318	—		Meta	48.2	100%	0%	0.2	90.3	131.1K	dominated
318	—		Meta	48.2	100%	0%	0	90.3	128K	dominated
320	—		MiniMax	47.8	100%	0%	0	—	—	dominated
320	—	non-reasoning reasoning	Amazon	47.8	100%	0%	3.4	124.2	—	dominated
320	—		Alibaba	47.8	100%	0%	0.8	—	—	dominated
323	—	minimal reasoning	OpenAI	47.2	100%	0%	0.7	—	—	dominated
323	—	low reasoning	OpenAI	47.2	100%	0%	0.1	198.5	—	dominated
323	—		Meta	47.2	100%	0%	0.4	104.8	1.0M	dominated
323	—		Alibaba	47.2	100%	0%	1.2	—	262.1K	dominated
327	—		DeepSeek	46.7	100%	0%	0.5	—	—	dominated
327	—	high reasoning	MBZUAI Institute of Foundation Models	46.7	100%	0%	0	—	—	dominated
327	—		NVIDIA	46.7	100%	0%	0.1	259.4	—	dominated
330	—	non-reasoning reasoning	Google	46.1	100%	0%	0.9	—	1.0M	dominated
330	—		InclusionAI	46.1	100%	0%	0.2	82.8	262.1K	dominated
330	—		Upstage	46.1	100%	0%	0.3	15	128K	dominated
330	—		Zyphra	46.1	100%	0%	0	—	—	dominated
334	—		OpenAI	45.7	100%	0%	0	—	—	dominated
335	—		Alibaba	45.5	100%	0%	0.9	194.4	262.1K	dominated
336	—		OpenAI	45.2	100%	0%	0	—	—	dominated
336	—		Alibaba	45.2	100%	0%	0.9	—	160K	dominated
336	—		Trillion Labs	45.2	100%	0%	0	—	—	dominated
339	—		Google	44.9	100%	0%	0	—	—	dominated
340	—		NVIDIA	44.5	100%	0%	1.2	36.6	131.1K	dominated
340	—		Alibaba	44.5	100%	0%	2.6	—	—	dominated
340	—		Alibaba	44.5	100%	0%	0.7	—	32.8K	dominated
343	—		Google	44.1	100%	0%	0	—	—	dominated
343	—		Alibaba	44.1	100%	0%	0.8	—	—	dominated
345	—	non-reasoning reasoning	Google	43.9	100%	0%	0.2	29.7	—	dominated
346	—	non-reasoning reasoning	Google	43.7	100%	0%	0.2	—	1.0M	dominated
347	—		InclusionAI	43.5	100%	0%	0	—	—	dominated
347	—		Motif Technologies	43.5	100%	0%	0	—	—	dominated
349	—		Amazon	43.2	100%	0%	5	65.3	—	dominated
350	—	medium reasoning	Mistral	42.8	100%	0%	0	—	—	dominated
350	—		Meta	42.8	100%	0%	0.4	31.2	131.1K	dominated
350	—		Mistral	42.8	100%	0%	0.8	—	131.1K	dominated
350	—		Upstage	42.8	100%	0%	0	—	—	dominated
354	—	medium reasoning	Mistral	42.1	100%	0%	0	—	131.1K	dominated
354	—	medium reasoning	MBZUAI Institute of Foundation Models	42.1	100%	0%	0	—	—	dominated
354	—		NVIDIA	42.1	100%	0%	0.4	93.4	—	dominated
354	—	non-reasoning reasoning	Mistral	42.1	100%	0%	0.3	145.3	—	dominated
354	—		Trillion Labs	42.1	100%	0%	0	—	—	dominated
359	—		Anthropic	41.4	100%	0%	0	—	200K	dominated
359	—		Google	41.4	100%	0%	0	—	—	dominated
359	—		OpenAI	41.4	100%	0%	0	—	—	dominated
362	—		Google	41.0	100%	0%	0	105.1	—	dominated
362	—		NVIDIA	41.0	100%	0%	0	—	—	dominated
364	—		OpenBMB	40.7	100%	0%	0	—	—	dominated
364	—		Alibaba	40.7	100%	0%	0	—	—	dominated
366	—	high reasoning	Sarvam	40.5	100%	0%	0.1	—	—	dominated
367	—		Celeris	39.9	100%	0%	3	2,116.3	—	dominated
367	—		Anthropic	39.9	100%	0%	30	—	—	dominated
367	—		Mistral	39.9	100%	0%	0	—	—	dominated
367	—		Google	39.9	100%	0%	0	—	—	dominated
367	—		Meta	39.9	100%	0%	0	169.2	16.4K	dominated
367	—	non-reasoning reasoning	Amazon	39.9	100%	0%	0.9	214.7	—	dominated
373	—	non-reasoning reasoning	Google	39.2	100%	0%	0	—	—	dominated
373	—	non-reasoning reasoning	OpenBMB	39.2	100%	0%	0	—	—	dominated
373	—		Perplexity	39.2	100%	0%	0	—	—	dominated
376	—		Alibaba	38.8	100%	0%	2.6	—	—	dominated
377	—		Google	38.7	100%	0%	0.2	—	—	dominated
378	—		Mistral	38.5	100%	0%	0.8	94.6	—	dominated
379	—		OpenAI	38.3	100%	0%	4.4	—	128K	dominated
380	—		Mistral	38.0	100%	0%	0.2	81.3	—	dominated
380	—		Nanbeige	38.0	100%	0%	0	—	—	dominated
380	—		Alibaba	38.0	100%	0%	1.2	—	131.1K	dominated
383	—		DeepSeek	37.6	100%	0%	0	—	32.8K	dominated
383	—	non-reasoning reasoning	Z AI	37.6	100%	0%	0.4	—	131.1K	dominated
385	—		AI21 Labs	37.3	100%	0%	3.5	57.4	256K	dominated
385	—	non-reasoning reasoning	Alibaba	37.3	100%	0%	1.2	—	—	dominated
387	—		Google	36.9	100%	0%	0	—	—	dominated
387	—		Mistral	36.9	100%	0%	0	—	—	dominated
389	—		LG AI Research	36.5	100%	0%	0	—	—	dominated
389	—		Mistral	36.5	100%	0%	0.2	—	—	dominated
389	—		Alibaba	36.5	100%	0%	0.7	—	—	dominated
392	—	non-reasoning reasoning	Amazon	36.2	100%	0%	0.9	—	—	dominated
393	—		DeepSeek	36.0	100%	0%	0	—	—	dominated
393	—		Alibaba	36.0	100%	0%	1.3	—	—	dominated
395	—	max reasoning	Alibaba	35.7	100%	0%	0	—	—	dominated
396	—		Google	35.2	100%	0%	0	—	—	dominated
396	—		Nous Research	35.2	100%	0%	0.2	91.2	—	dominated
396	—		Meta	35.2	100%	0%	0.3	82	327.7K	dominated
396	—		Alibaba	35.2	100%	0%	0.4	—	131.1K	dominated
396	—	non-reasoning reasoning	Upstage	35.2	100%	0%	0	—	—	dominated
401	—		Anthropic	34.7	100%	0%	6	—	—	dominated
401	—		DeepSeek	34.7	100%	0%	0.8	—	131.1K	dominated
403	—		DeepSeek	34.3	100%	0%	0	—	—	dominated
403	—		TII UAE	34.3	100%	0%	0	—	—	dominated
403	—		Mistral	34.3	100%	0%	3	—	65.5K	dominated
406	—		Google	33.7	100%	0%	0	—	—	dominated
406	—		InclusionAI	33.7	100%	0%	0.2	—	—	dominated
406	—		Meta	33.7	100%	0%	0	52.1	131.1K	dominated
406	—		Meta	33.7	100%	0%	0	52.1	131.1K	dominated
410	—		OpenAI	33.0	100%	0%	0.2	—	1.0M	dominated
410	—		OpenAI	33.0	100%	0%	4.4	—	128K	dominated
410	—		Alibaba	33.0	100%	0%	0.5	—	—	dominated
410	—		Alibaba	33.0	100%	0%	0.4	107.5	—	dominated
414	—		Perplexity	32.5	100%	0%	1	—	127.1K	dominated
414	—		StepFun	32.5	100%	0%	0	—	—	dominated
416	—		Meta	32.3	100%	0%	0.6	75.5	—	dominated
417	—		Mistral	32.0	100%	0%	0	—	131.1K	dominated
417	—		Alibaba	32.0	100%	0%	0.8	—	—	dominated
417	—		Perplexity	32.0	100%	0%	6	—	200K	dominated
420	—		Alibaba	31.6	100%	0%	0	—	—	dominated
421	—		Z AI	31.2	100%	0%	0.9	—	—	dominated
421	—		NVIDIA	31.2	100%	0%	0.9	51.4	—	dominated
421	—		Mistral	31.2	100%	0%	0	—	—	dominated
421	—		Alibaba	31.2	100%	0%	0.4	—	—	dominated
425	—		Baidu	30.4	100%	0%	0.5	—	123K	dominated
425	—		Nous Research	30.4	100%	0%	1.5	36.3	—	dominated
425	—		Mistral	30.4	100%	0%	0.2	91.5	—	dominated
425	—		NVIDIA	30.4	100%	0%	0.3	79.4	—	dominated
425	—		OpenAI	30.4	100%	0%	0.8	103.6	16.4K	dominated
425	—		Upstage	30.4	100%	0%	0	—	—	dominated
431	—	non-reasoning reasoning	Google	29.7	100%	0%	0	102.9	—	dominated
431	—		IBM	29.7	100%	0%	0	—	—	dominated
431	—		Meta	29.7	100%	0%	0.6	46	8.2K	dominated
434	—		Google	29.2	100%	0%	0	—	1.0M	dominated
434	—	non-reasoning reasoning	Nous Research	29.2	100%	0%	1.5	36.2	—	dominated
434	—		NVIDIA	29.2	100%	0%	0.1	245.7	—	dominated
437	—	non-reasoning reasoning	NVIDIA	28.7	100%	0%	0.4	99.9	—	dominated
437	—		Meta	28.7	100%	0%	0	77.4	131.1K	dominated
437	—		NVIDIA	28.7	100%	0%	0	—	—	dominated
440	—		Google	28.1	100%	0%	0	—	—	dominated
440	—		OpenAI	28.1	100%	0%	7.5	—	128K	dominated
440	—	low reasoning	MBZUAI Institute of Foundation Models	28.1	100%	0%	0	—	—	dominated
440	—	non-reasoning reasoning	Alibaba	28.1	100%	0%	1.2	—	—	dominated
444	—		Kimi	27.5	100%	0%	0	—	—	dominated
444	—		Meta	27.5	100%	0%	4.4	—	—	dominated
444	—		NVIDIA	27.5	100%	0%	0	—	—	dominated
444	—	non-reasoning reasoning	NVIDIA	27.5	100%	0%	0	—	—	dominated
448	—		Alibaba	27.0	100%	0%	0	—	—	dominated
448	—		Alibaba	27.0	100%	0%	0.3	—	131.1K	dominated
450	—		Anthropic	26.5	100%	0%	6	—	—	dominated
450	—		Liquid AI	26.5	100%	0%	0	342	—	dominated
450	—		Alibaba	26.5	100%	0%	0.7	—	—	dominated
450	—		Allen Institute for AI	26.5	100%	0%	0	—	—	dominated
454	—		OpenAI	26.0	100%	0%	0	—	—	dominated
454	—		InclusionAI	26.0	100%	0%	0.2	—	—	dominated
456	—		Allen Institute for AI	25.7	100%	0%	0	—	65.5K	dominated
456	—		Mistral	25.7	100%	0%	0	—	131.1K	dominated
458	—		Google	25.3	100%	0%	0	—	—	dominated
458	—	minimal reasoning	OpenAI	25.3	100%	0%	0.1	—	—	dominated
458	—		SpaceXAI	25.3	100%	0%	0	—	—	dominated
461	—		OpenAI	24.9	100%	0%	15	—	128K	dominated
461	—		Alibaba	24.9	100%	0%	0	—	—	dominated
463	—	non-reasoning reasoning	Upstage	24.6	100%	0%	0	—	—	dominated
464	—		Cohere	24.4	100%	0%	4.4	53.6	256K	dominated
464	—		Amazon	24.4	100%	0%	1.4	—	—	dominated
466	—		Meta	24.0	100%	0%	0.1	—	—	dominated
466	—		NVIDIA	24.0	100%	0%	1.2	78.3	—	dominated
466	—		Alibaba	24.0	100%	0%	0	—	—	dominated
469	—		SpaceXAI	23.6	100%	0%	0	—	—	dominated
469	—		Alibaba	23.6	100%	0%	0	—	—	dominated
471	—		Google	23.2	100%	0%	0	—	—	dominated
471	—	non-reasoning reasoning	NVIDIA	23.2	100%	0%	0.1	218	256K	dominated
471	—	non-reasoning reasoning	NVIDIA	23.2	100%	0%	0.1	176.5	128K	dominated
474	—		Mistral	22.8	100%	0%	3	—	131.1K	dominated
475	—		Alibaba	22.6	100%	0%	0	—	—	dominated
475	—		Alibaba	22.6	100%	0%	0	—	—	dominated
477	—	non-reasoning reasoning	Z AI	22.2	100%	0%	0.9	—	65.5K	dominated
477	—		OpenAI	22.2	100%	0%	37.5	—	8.2K	dominated
477	—	non-reasoning reasoning	Alibaba	22.2	100%	0%	0.6	—	—	dominated
480	—	non-reasoning reasoning	Google	21.5	100%	0%	0.2	—	1.0M	dominated
480	—		OpenAI	21.5	100%	0%	0.3	—	128K	dominated
480	—	non-reasoning reasoning	Nous Research	21.5	100%	0%	0.2	86	—	dominated
480	—		Mistral	21.5	100%	0%	0.2	—	—	dominated
480	—		Amazon	21.5	100%	0%	0.1	—	—	dominated
485	—		DeepSeek	20.7	100%	0%	0	—	—	dominated
485	—		Meta	20.7	100%	0%	0.6	—	—	dominated
485	—		Mistral	20.7	100%	0%	0.1	295.3	—	dominated
485	—	non-reasoning reasoning	Alibaba	20.7	100%	0%	0.4	—	—	dominated
485	—	non-reasoning reasoning	Alibaba	20.7	100%	0%	0	—	—	dominated
490	—		IBM	20.2	100%	0%	0.1	116.2	131.1K	dominated
491	—		DeepSeek	19.9	100%	0%	0	—	—	dominated
491	—		Google	19.9	100%	0%	0	—	—	dominated
491	—	high reasoning	Sarvam	19.9	100%	0%	0	—	—	dominated
494	—		Allen Institute for AI	19.6	100%	0%	0	—	65.5K	dominated
495	—		DeepSeek	19.1	100%	0%	0	—	—	dominated
495	—	non-reasoning reasoning	Google	19.1	100%	0%	0	—	—	dominated
495	—		Meta	19.1	100%	0%	0	82.2	8.2K	dominated
495	—		Mistral	19.1	100%	0%	0.3	—	32.8K	dominated
495	—		Allen Institute for AI	19.1	100%	0%	0.2	—	65.5K	dominated
500	—		Google	18.3	100%	0%	0	—	—	dominated
500	—		Meta	18.3	100%	0%	0.1	139.8	60K	dominated
500	—		Alibaba	18.3	100%	0%	0.1	—	131.1K	dominated
500	—		Perplexity	18.3	100%	0%	0	—	—	dominated
500	—		Reka AI	18.3	100%	0%	0.4	—	—	dominated
505	—		Meta	17.7	100%	0%	0	—	—	dominated
505	—		Upstage	17.7	100%	0%	0.2	—	—	dominated
507	—	non-reasoning reasoning	LG AI Research	17.2	100%	0%	0	—	—	dominated
507	—		SpaceXAI	17.2	100%	0%	0	—	—	dominated
507	—		Microsoft	17.2	100%	0%	0	44.9	—	dominated
507	—		Alibaba	17.2	100%	0%	0	—	—	dominated
511	—	non-reasoning reasoning	Alibaba	16.8	100%	0%	0	—	—	dominated
512	—		Google	16.5	100%	0%	0	—	—	dominated
512	—		Google	16.5	100%	0%	0	—	—	dominated
512	—		Alibaba	16.5	100%	0%	0	—	—	dominated
515	—	non-reasoning reasoning	Nous Research	16.1	100%	0%	0	—	32.8K	dominated
515	—		AI21 Labs	16.1	100%	0%	3.5	56.2	—	dominated
517	—		IBM	15.8	100%	0%	0.1	448.9	—	dominated
518	—		DeepSeek	15.3	100%	0%	0	—	—	dominated
518	—		Nous Research	15.3	100%	0%	0.7	—	65.5K	dominated
518	—		AI21 Labs	15.3	100%	0%	3.5	—	—	dominated
518	—	non-reasoning reasoning	Alibaba	15.3	100%	0%	0.3	—	—	dominated
518	—		Alibaba	15.3	100%	0%	0.4	107.9	—	dominated
523	—		AI21 Labs	14.8	100%	0%	3.5	—	—	dominated
523	—		Allen Institute for AI	14.8	100%	0%	0	—	—	dominated
525	—		Google	14.4	100%	0%	0	—	—	dominated
525	—		Liquid AI	14.4	100%	0%	0	—	—	dominated
525	—		Microsoft	14.4	100%	0%	0.2	41.9	16.4K	dominated
528	—		Anthropic	13.8	100%	0%	6	—	—	dominated
528	—		IBM	13.8	100%	0%	0	—	—	dominated
528	—		Mistral	13.8	100%	0%	0.3	—	—	dominated
528	—		Amazon	13.8	100%	0%	0.1	265.4	—	dominated
532	—		Google	13.1	100%	0%	0	—	—	dominated
532	—		Google	13.1	100%	0%	0	—	—	dominated
532	—	non-reasoning reasoning	NVIDIA	13.1	100%	0%	0.3	159.7	128K	dominated
532	—		Microsoft	13.1	100%	0%	0	—	—	dominated
536	—		Microsoft	12.6	100%	0%	0	26.6	—	dominated
536	—		Alibaba	12.6	100%	0%	0	—	32.8K	dominated
538	—		Mistral	12.3	100%	0%	0	—	—	dominated
538	—		Mistral	12.3	100%	0%	6	—	128K	dominated
540	—		Meta	12.1	100%	0%	0.1	—	—	dominated
541	—		Meta	11.8	100%	0%	0	—	—	dominated
541	—		OpenBMB	11.8	100%	0%	0	—	—	dominated
543	—		AI21 Labs	11.3	100%	0%	0	—	—	dominated
543	—		Alibaba	11.3	100%	0%	0	—	—	dominated
543	—		Alibaba	11.3	100%	0%	0	—	—	dominated
543	—		Reka AI	11.3	100%	0%	0.4	96.9	65.5K	dominated
547	—		Allen Institute for AI	10.9	100%	0%	0	—	65.5K	dominated
548	—		Anthropic	10.6	100%	0%	0	—	—	dominated
548	—		Anthropic	10.6	100%	0%	0.5	—	200K	dominated
548	—		Allen Institute for AI	10.6	100%	0%	0	—	—	dominated
551	—		InclusionAI	10.2	100%	0%	0	—	—	dominated
551	—		Allen Institute for AI	10.2	100%	0%	0	—	—	dominated
553	—		DeepSeek	10.0	100%	0%	0	—	—	dominated
554	—		Anthropic	9.5	100%	0%	0	—	—	dominated
554	—		DeepSeek	9.5	100%	0%	0	—	—	dominated
554	—		OpenAI	9.5	100%	0%	0.8	—	—	dominated
554	—	medium reasoning	Mistral	9.5	100%	0%	3	—	—	dominated
554	—		Mistral	9.5	100%	0%	0.3	—	—	dominated
559	—		Meta	9.0	100%	0%	1.2	—	—	dominated
560	—		Snowflake	8.6	100%	0%	0	—	—	dominated
560	—		Liquid AI	8.6	100%	0%	0	—	—	dominated
560	—		Alibaba	8.6	100%	0%	0	—	—	dominated
563	—		Meta	8.2	100%	0%	0.3	2.5	—	dominated
563	—	non-reasoning reasoning	Alibaba	8.2	100%	0%	0	—	—	dominated
565	—		Google	8.0	100%	0%	0	—	—	dominated
566	—		DeepSeek	7.7	100%	0%	0	—	—	dominated
566	—		Google	7.7	100%	0%	0	—	—	dominated
568	—		Cohere	7.0	100%	0%	6	—	—	dominated
568	—		Databricks	7.0	100%	0%	0	—	—	dominated
568	—		DeepSeek	7.0	100%	0%	0	—	—	dominated
568	—		Meta	7.0	100%	0%	0	—	—	dominated
568	—		Meta	7.0	100%	0%	0	—	—	dominated
568	—		OpenChat	7.0	100%	0%	0	—	—	dominated
568	—		Sarvam	7.0	100%	0%	0	—	—	dominated
575	—		LG AI Research	6.4	100%	0%	0	—	—	dominated
576	—	non-reasoning reasoning	LG AI Research	6.1	100%	0%	0	—	—	dominated
576	—		Allen Institute for AI	6.1	100%	0%	0.1	—	65.5K	dominated
578	—		IBM	5.5	100%	0%	0	—	—	dominated
578	—		AI21 Labs	5.5	100%	0%	0.3	—	—	dominated
578	—		AI21 Labs	5.5	100%	0%	0	—	—	dominated
578	—		Liquid AI	5.5	100%	0%	0	—	—	dominated
578	—		Liquid AI	5.5	100%	0%	0	—	—	dominated
578	—		Liquid AI	5.5	100%	0%	0	—	—	dominated
584	—		AI21 Labs	4.8	100%	0%	0.3	—	—	dominated
584	—		Alibaba	4.8	100%	0%	0	—	—	dominated
586	—		Swiss AI Initiative	4.3	100%	0%	1.3	—	—	dominated
586	—		Google	4.3	100%	0%	0	—	—	dominated
586	—		IBM	4.3	100%	0%	0	—	—	dominated
586	—		Mistral	4.3	100%	0%	0.5	—	32.8K	dominated
590	—	non-reasoning reasoning	Nous Research	3.9	100%	0%	0	—	—	dominated
591	—		Anthropic	3.3	100%	0%	0	—	—	dominated
591	—		Cohere	3.3	100%	0%	0.8	—	—	dominated
591	—		IBM	3.3	100%	0%	0	—	—	dominated
591	—		Meta	3.3	100%	0%	0	—	—	dominated
591	—		Mistral	3.3	100%	0%	0.3	—	32.8K	dominated
591	—		Alibaba	3.3	100%	0%	0	—	—	dominated
597	—		Allen Institute for AI	2.8	100%	0%	0	—	—	dominated
598	—	non-reasoning reasoning	IBM	2.5	100%	0%	0.1	—	—	dominated
598	—		Liquid AI	2.5	100%	0%	0	—	32.8K	dominated
600	—	non-reasoning reasoning	Alibaba	2.3	100%	0%	0	—	—	dominated
601	—		Alibaba	2.1	100%	0%	0	—	—	dominated
602	—		Google	1.9	100%	0%	0.1	—	—	dominated
602	—		Meta	1.9	100%	0%	0.1	—	—	dominated
604	—		Google	1.5	100%	0%	0	—	—	dominated
604	—		Liquid AI	1.5	100%	0%	0	—	—	dominated
604	—		Meta	1.5	100%	0%	0	—	—	dominated
607	—		Swiss AI Initiative	0.6	100%	0%	0.1	—	—	dominated
607	—		Google	0.6	100%	0%	0	—	—	dominated
607	—		Google	0.6	100%	0%	0	—	—	dominated
607	—		IBM	0.6	100%	0%	0	—	—	dominated
607	—		IBM	0.6	100%	0%	0	—	—	dominated
607	—		Liquid AI	0.6	100%	0%	0	373.9	—	dominated
607	—	non-reasoning reasoning	Alibaba	0.6	100%	0%	0	—	—	dominated
607	—		Cohere	0.6	100%	0%	0	130	—	dominated

331 excludedby population, evidence, or constraints

DeepSeek-OCR
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Grok Voice Agent
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Cogito v2.1 (Reasoning)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mi:dm K 2.5 Pro Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Doubao-Seed-1.8
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GPT-4o mini Realtime (Dec '24)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GPT-4o Realtime (Dec '24)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GPT-3.5 Turbo (0613)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Aurora Alpha
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Free Models Router
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
StepFun: Step 3.5 Flash (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Trinity Large Preview (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Upstage: Solar Pro 3 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MiniMax: MiniMax M2-her
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Writer: Palmyra X5
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
LiquidAI: LFM2.5-1.2B-Thinking (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
LiquidAI: LFM2.5-1.2B-Instruct (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT Audio
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT Audio Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AllenAI: Molmo2 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ByteDance Seed: Seed 1.6 Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ByteDance Seed: Seed 1.6
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small Creative
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.2 Chat
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.2 Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Devstral 2 2512
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Relace: Relace Search
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nex AGI: DeepSeek V3.1 Nex N1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
EssentialAI: Rnj 1 Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Body Builder (beta)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.1-Codex-Max
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Ministral 3 14B 2512
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Ministral 3 8B 2512
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Ministral 3 3B 2512
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Large 3 2512
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Trinity Mini (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TNG: R1T Chimera (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Nano Banana Pro (Gemini 3 Pro Image Preview)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3 Pro Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Deep Cogito: Cogito v2.1 671B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.1 Chat
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Kwaipilot: KAT-Coder-Pro V1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Perplexity: Sonar Pro Search
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Voxtral Small 24B 2507
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: gpt-oss-safeguard-20b
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
LiquidAI: LFM2-2.6B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
IBM: Granite 4.0 Micro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Image Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 VL 8B Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Image
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: o3 Deep Research
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: o4 Mini Deep Research
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NVIDIA: Llama 3.3 Nemotron Super 49B V1.5
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu: ERNIE 4.5 21B A3B Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Flash Image (Nano Banana)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 VL 30B A3B Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: DeepSeek V3.2 Exp
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TheDrummer: Cydonia 24B V4.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Relace: Relace Apply 3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 VL 235B A22B Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Coder Plus
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Tongyi DeepResearch 30B A3B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Coder Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenGVLab: InternVL3 78B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Next 80B A3B Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meituan: LongCat Flash Chat
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen Plus 0728
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 30B A3B Thinking 2507
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nous: Hermes 4 70B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nous: Hermes 4 405B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: DeepSeek V3.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o Audio
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu: ERNIE 4.5 21B A3B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu: ERNIE 4.5 VL 28B A3B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Chat
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Codestral 2508
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 235B A22B Thinking 2507
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Z.ai: GLM 4 32B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Coder 480B A35B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ByteDance: UI-TARS 7B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 235B A22B Instruct 2507
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Switchpoint Router
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Venice: Uncensored (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 3n 2B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Tencent: Hunyuan A13B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TNG: DeepSeek R1T2 Chimera (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Morph: Morph V3 Large
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Morph: Morph V3 Fast
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu: ERNIE 4.5 VL 424B A47B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Inception: Mercury
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small 3.2 24B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MiniMax: MiniMax M1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 3 Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Pro Preview 06-05
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: R1 0528 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 3n 4B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Pro Preview 05-06
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Spotlight
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Maestro Reasoning
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Virtuoso Large
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Coder Large
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Inception: Mercury Coder
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 4B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meta: Llama Guard 4 12B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 30B A3B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 14B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 32B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 235B A22B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TNG: DeepSeek R1T Chimera (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: o4 Mini High
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
EleutherAI: Llemma 7b
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AlfredPros: CodeLLaMa 7B Instruct Solidity
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 3 Mini Beta
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 3 Beta
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NVIDIA: Llama 3.1 Nemotron Ultra 253B v1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen2.5 VL 32B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: DeepSeek V3 0324
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small 3.1 24B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AllenAI: Olmo 2 32B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 3 4B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 3 12B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o-mini Search Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o Search Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 3 27B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TheDrummer: Skyfall 36B V2
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Perplexity: Sonar Deep Research
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Llama Guard 3 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.0 Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen VL Plus
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-1.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-1.0-Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-RP 1.0 (8B)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen VL Max
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen2.5 VL 72B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen-Plus
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen-Max
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small 3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MiniMax: MiniMax-01
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sao10K: Llama 3.1 70B Hanami x1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sao10K: Llama 3.3 Euryale 70B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Cohere: Command R7B (12-2024)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o (2024-11-20)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral Large 2411
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen2.5 Coder 32B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
SorcererLM 8x22B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TheDrummer: UnslopNemo 12B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Magnum v4 72B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude 3.5 Sonnet
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen2.5 7B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Inflection: Inflection 3 Pi
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Inflection: Inflection 3 Productivity
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
TheDrummer: Rocinante 12B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen2.5 72B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NeverSleep: Lumimaid v0.2 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Pixtral 12B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Cohere: Command R (08-2024)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Cohere: Command R+ (08-2024)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sao10K: Llama 3.1 Euryale 70B v2.2
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen2.5-VL 7B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nous: Hermes 3 405B Instruct (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: ChatGPT-4o
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sao10K: Llama 3 8B Lunaris
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meta: Llama 3.1 405B (base)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meta: Llama 3.1 405B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Nemo
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o-mini (2024-07-18)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 2 27B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 2 9B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sao10k: Llama 3 Euryale 70B v2.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NousResearch: Hermes 2 Pro - Llama-3 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral 7B Instruct v0.3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meta: LlamaGuard 2 8B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
WizardLM-2 8x22B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4 Turbo Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral 7B Instruct v0.2
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Noromaid 20B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Goliath 120B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Auto Router
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4 Turbo (older v1106)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-3.5 Turbo Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral 7B Instruct v0.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-3.5 Turbo 16k
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mancer: Weaver (alpha)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ReMM SLERP 13B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MythoMax 13B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4 (older v0314)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen Plus 0728 (thinking)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Coder 480B A35B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 4B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small 3.1 24B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nous: Hermes 3 405B Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.5 Plus 2026-02-15
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Amazon: Nova 2 Lite
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Amazon: Nova Premier 1.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Amazon: Nova Lite 1.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Amazon: Nova Micro 1.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Amazon: Nova Pro 1.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-2.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.1 Pro Preview Custom Tools
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.5-Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
LiquidAI: LFM2-24B-A2B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ByteDance Seed: Seed-2.0-Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.4 Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.4
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.3 Chat
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GPT-5.4 Pro (xhigh)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 4.20 Multi-Agent Beta
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 4.20 Beta
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Hunter Alpha
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Healer Alpha
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
ByteDance Seed: Seed-2.0-Lite
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Mistral: Mistral Small 4
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Reka Edge
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Gemini 3 Deep Think
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Lyria 3 Pro Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Lyria 3 Clip Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.6 Plus Preview (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok 4.20 Multi-Agent
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Reka Edge
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Arcee AI: Trinity Large Thinking
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.6 Plus (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 4 31B
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.7
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Elephant
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.6 (Fast)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemma 4 26B A4B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meta: Llama Guard 4 12B (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Opus 4.7
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Mythos Preview
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nanonets OCR-3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nanonets OCR2+
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GLM-OCR
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic Claude Haiku Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI GPT Mini Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google Gemini Pro Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MoonshotAI Kimi Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google Gemini Flash Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic Claude Sonnet Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI GPT Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.5 Plus 2026-04-20
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.6 Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.5 Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.4 Image 2
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Pareto Code Router
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu: Qianfan-OCR-Fast (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
EXAONE 4.5 33B (Non-reasoning)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NVIDIA: Nemotron 3 Nano Omni (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Poolside: Laguna XS.2 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Poolside: Laguna M.1 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
inclusionAI: Ling-2.6-flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MiniMax: MiniMax M2.5 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Z.ai: GLM 4.6
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3 Next 80B A3B Instruct (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MoonshotAI: Kimi K2 0905
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: gpt-oss-120b (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: gpt-oss-20b (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Z.ai: GLM 4.5 Air (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude 3.7 Sonnet (thinking)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-4o
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Owl Alpha
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
GPT-5.5 Pro (xhigh)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT Chat Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Microsoft: Phi 4 Mini Instruct
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Baidu Qianfan: CoBuddy (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.1 Flash Lite
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
inclusionAI: Ling-2.6-1T
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
inclusionAI: Ring-2.6-1T (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.7 (Fast)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Perceptron: Perceptron Mk1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: DeepSeek V4 Flash (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok Build 0.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MoonshotAI: Kimi K2.6 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.8 (Fast)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
StepFun: Step 3.7 Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MiniMax: MiniMax M3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenRouter: Fusion
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
NVIDIA: Nemotron 3.5 Content Safety (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nex AGI: Nex-N2-Pro (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Fable Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MoonshotAI: Kimi K2.7 Code
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Z.ai: GLM 5.2
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Nano Banana 2 (Gemini 3.1 Flash Image)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Nano Banana Pro (Gemini 3 Pro Image)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Sakana: Fugu Ultra
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Sonnet 5 (Adaptive Reasoning, High Effort)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Sonnet 5 (Adaptive Reasoning, Medium Effort)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Sonnet 5 (Adaptive Reasoning, Low Effort)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Poolside: Laguna XS 2.1 (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Tencent: Hy3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Nex AGI: Nex-N2-Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-3.0-Mini
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
AionLabs: Aion-3.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
xAI: Grok Latest
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.6 Luna Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.6 Terra Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.6 Sol Pro
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Opus 5 (Fast)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Ling-3.0-flash (free)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Poolside: Laguna S 2.1
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Meituan: LongCat 2.0
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Auto Router (Beta)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
MoonshotAI: Kimi K3
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Kwaipilot: KAT-Coder-Air V2.5
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Kwaipilot: KAT-Coder-Pro V2.5
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Qwen: Qwen3.7 Flash
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Claude Opus 5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Sonnet 5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Fable 5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.8 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.6 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.2 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Anthropic: Claude Opus 4.5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5.1 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Mini (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
OpenAI: GPT-5 Nano (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.6 Flash (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.5 Flash Lite (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.5 Flash (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 3.1 Pro Preview (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Flash Lite (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Flash (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
Google: Gemini 2.5 Pro (batch)
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.
DeepSeek: DeepSeek V4 Flash 0731
- Too few Source Benchmarks contribute to this score.
- Too few contributions are directly observed.

Why this score?

Why Claude Opus 5 (Adaptive Reasoning, Max Effort) scored 100.0

#1Rank

100.0%Coverage

1Observed

0Predicted

Intelligence Index

Observed

+100.0score points

Raw: 60.70
Normalized: 100.00
Configured weight: 100.0%
Effective weight: 100.0%

artificial_analysis

Source date 2026-07-31 · recorded 2026-07-31

Methodology & snapshot

Formula

Intelligence Index × 100.0%higher · percentile

Missing data: reweight available
Evidence: observed only
Population: Population: all models
Coverage gate: 0% · 1 observed · 1 contributing
Snapshot: Captured 2026-07-31 · 945 models · 83992 receipts

Integrity warnings1

Intelligence Index has moderate contamination riskinfo
Weighting is set by Artificial Analysis and encodes subjective design choices even when documented; strong performance on one high-value workload can be hidden by weaker results in other categories.
Remove it, lower its weight, or document why its contamination risk is acceptable.

Sensitivity

Perturb every configured weight by ±20% to test whether small judgment changes reorder the result.