AI Company Leaderboard · The Aggregate

Company leaderboard

Ranked by the strongest currently ranked model. How these rankings work.

Rank	Company	Strongest model	Best model rating	Ranked models
1	Anthropic	Claude Mythos 5.1 (Private)	2054 ± 28	21
2	Google	Gemini 4 Argon (Private)	2015 ± 18	65
3	OpenAI	GPT-6 Astra	2011 ± 5	75
4	xAI	Grok 4.7	1894 ± 9	20
5	Meta	Muse Spark 1.3	1893 ± 7	72
6	Alibaba	Qwen 3.8 Max (0902)	1881 ± 13	182
7	StepFun	Step 5 Preview	1880 ± 16	6
8	Moonshot	Kimi K3	1875 ± 4	10
9	DeepSeek	DeepSeek V4.1 Flash	1860 ± 7	39
9	Zhipu	GLM-5.3	1860 ± 6	33
11	Fireworks AI	Ember-1 (Low coverage)	1858 ± 32	1
12	Xiaomi	MiMo-V2.6-Pro	1852 ± 12	7
13	Tencent	Hy4 preview	1845 ± 11	8
14	ByteDance	Seed 2.1 Pro	1818 ± 9	20
15	InclusionAI	Ling 3.1 Flash	1803 ± 39	10
16	Apodex	Apodex 1.1	1798 ± 28	1
17	Perplexity	Sonar Deep Research (Low coverage)	1797 ± 32	3
18	Cursor	Composer 2.5 (Low coverage)	1777 ± 21	1
19	Thinking Machines	Inkling	1763 ± 8	2
20	MiniMax	MiniMax-M3	1754 ± 4	11
21	Nex AGI	Nex N2 Pro	1751 ± 16	2
22	Upstage	Solar Open2 250B	1733 ± 27	8
23	Meituan	LongCat-Flash-Thinking-2601	1717 ± 21	5
24	Motif	Motif 3	1712 ± 21	1
25	Kuaishou	KAT Coder Pro V2	1705 ± 49	2
25	Skywork	Skywork-Reward-Gemma-2-27B-v0.2 (Low coverage)	1705 ± 31	1
27	Baidu	Ernie 5.1	1702 ± 20	9
28	Applied Innovation Center	AIC-1	1699 ± 34	1
29	NVIDIA	NVIDIA-Nemotron-3-Super-120B-A12B-FP8	1694 ± 18	20
30	Mistral	Mistral-Large-3-675B-Instruct-2512 (Low coverage)	1687 ± 21	60
31	Poolside	Laguna S 2.1	1663 ± 20	4
32	Shanghai AI Lab	Intern-S1 (Low coverage)	1659 ± 19	29
33	SK Telecom	A.X-K2	1658 ± 20	4
34	SenseTime	SenseNova V6 Reasoner (Low coverage)	1655 ± 40	1
35	Cohere	command-a-reasoning-08-2025 (Low coverage)	1651 ± 27	24
36	Arcee AI	Arcee Trinity Large	1648 ± 17	9
36	Nous Research	Hermes 4 405B	1648 ± 19	17
38	iFlytek	Spark X2 (Low coverage)	1646 ± 28	3
39	KAUST	Hala-9B	1632 ± 34	1
40	LG AI	K-EXAONE	1631 ± 14	8
41	Amazon	Nova 2.0 Pro Preview	1629 ± 48	11
42	T-Bank	T-pro-it-2.0 (Low coverage)	1628 ± 34	1
43	TIGER-Lab	Qwen2.5-32B-Instruct-CFT	1626 ± 27	3
44	Qihoo 360	360Zhinao2-O1.5	1624 ± 35	2
45	FreedomIntelligence	AceGPT-v2-70B-Chat	1621 ± 26	4
45	OpenBMB	MiniCPM-o-4.5	1621 ± 11	3
47	Inception (G42), Cerebras & MBZUAI	Jais-2-70B-Chat	1616 ± 31	2
48	Inception	Mercury 2.5	1614 ± 27	2
48	National Institute of Informatics	llm-jp-4-32B-a3B	1614 ± 20	8
50	Sarvam AI	Sarvam 105B	1611 ± 34	3
51	01.AI	Yi Lightning	1610 ± 18	23
52	IBM	Granite 4.2 30B	1606 ± 34	23
53	PrimeIntellect	INTELLECT-3	1605 ± 18	1
54	Institute of Science Tokyo	GPT-OSS-Swallow-20B-RL-v0.1	1604 ± 23	3
55	Allen AI	Llama-3.1-Tulu-3-70B	1596 ± 21	26
56	AXCXEPT	EZO-Qwen2.5-32B-Instruct	1592 ± 26	1
57	INSAIT	MamayLM-Gemma-3-12B-IT-v1.0	1583 ± 39	1
58	ABEJA	ABEJA-Qwen2.5-32B-Japanese-v1.0	1582 ± 20	1
59	rinna	qwq-bakeneko-32B	1581 ± 22	2
60	China Mobile	JT-MINI	1579 ± 50	1
61	AI21 Labs	jamba-large-1.7	1577 ± 13	6
62	CASIA & Wuhan AI	Taichu-VLR-7B	1575 ± 24	2
63	Return Zero	ko-gemma-2-9B-it (Low coverage)	1573 ± 19	1
64	ELYZA	ELYZA-Shortcut-1.0-Qwen-32B	1570 ± 19	1
64	UCSC-VLAA	VLAA-Thinker-Qwen2.5VL-7B	1570 ± 24	2
66	Sber	GigaChat 2 Max	1568 ± 30	9
66	SpeakLeash	Bielik-11B-v2.5-Instruct	1568 ± 20	7
68	AI Singapore	Llama 3.1 70B Cpt Sea Lionv3 Instruct	1567 ± 26	5
69	QCRI	Fanar-1-9B-Instruct	1565 ± 22	2
70	China Telecom	TeleChat2	1562 ± 45	1
70	Sea AI Lab	Sailor2-20B-Chat	1562 ± 21	3
72	Princeton NLP	gemma-2-9B-it-SimPO	1561 ± 16	48
72	Reka	Reka Flash 3	1561 ± 21	1
74	Sahabat-AI	gemma2-9B-cpt-sahabatai-v1-instruct (Low coverage)	1556 ± 21	2
75	Abacus.AI	Smaug-72B-v0.1 (Low coverage)	1555 ± 24	8
76	Axolotl	romulus-mistral-nemo-12B-simpo (Low coverage)	1552 ± 20	1
77	LLM360	K2 Think V2	1551 ± 27	1
78	A*STAR	MERaLiON-2-10B	1549 ± 17	1
79	UCLA	Gemma-2-9B-It-SPPO-Iter2 (Low coverage)	1548 ± 20	7
80	CyberAgent	DeepSeek-R1-Distill-Qwen-32B-Japanese	1547 ± 19	1
80	SB Intuitions	sarashina2-70B (Low coverage)	1547 ± 48	7
82	Inception (G42) & Cerebras	jais-adapted-70B-chat	1538 ± 24	4
83	Naver	HyperCLOVA X SEED Think (32B)	1537 ± 22	1
84	Lapa	lapa-v0.1.2-instruct	1535 ± 39	2
85	Cognitive Computations	dolphin-2.9.1-llama-3-70B (Low coverage)	1533 ± 31	13
86	Microsoft	Phi-3.5-MoE-instruct	1531 ± 16	21
87	University of Ljubljana	GaMS3-12B-Instruct	1527 ± 22	1
88	TII	Falcon-H1R-7B	1525 ± 39	10
89	Jon Durbin	Bagel-Hermes-34B-Slerp (Low coverage)	1523 ± 24	3
90	Linkbricks Horizon-AI	Linkbricks-Horizon-AI-Korean-Superb-22B	1521 ± 18	1
91	BAAI	Gemma2-9B-IT-Simpo-Infinity-Preference (Low coverage)	1515 ± 21	9
92	Tenyx	Llama3-TenyxChat-70B (Low coverage)	1514 ± 24	1
93	PLLuM	Llama-PLLuM-70B-chat (Low coverage)	1513 ± 23	6
94	SDAIA	ALLaM-7B-Instruct-preview	1507 ± 19	1
95	omlab	VLM-R1-3B-Math-0305	1503 ± 24	1
96	VAGOsolutions	Llama-3.1-SauerkrautLM-8B-Instruct (Low coverage)	1501 ± 20	5
97	KT	Midm-2.0-Base-Instruct	1498 ± 30	2
98	URSA-MATH	URSA-8B	1494 ± 22	1
99	Liquid AI	LFM2.5-VL-1.6B	1491 ± 27	12
100	Apertus	Apertus-70B-Instruct-2509	1486 ± 18	2
101	Valiant Labs	Llama 3 70B Fireplace (Low coverage)	1484 ± 24	4
102	OrionStar	OrionStar-Yi-34B-Chat	1482 ± 34	1
102	SILMA AI	SILMA-9B-Instruct-v1.0	1482 ± 24	1
104	Yanolja	EEVE-Korean-Instruct-10.8B-v1.0 (Low coverage)	1479 ± 18	3
105	HuggingFace	starchat2-15B-v0.1	1470 ± 21	14
106	Stanford CRFM	Marin 8B Instruct (Low coverage)	1467 ± 84	2
107	Preferred Elements	plamo-2-8B (Low coverage)	1466 ± 48	2
108	Nanbeige	Nanbeige4.1-3B	1465 ± 38	2
109	Argilla	notux-8x7B-v1 (Low coverage)	1461 ± 31	1
109	Lightblue	suzume-llama-3-8B-multilingual-orpo-borda-top75 (Low coverage)	1461 ± 18	5
111	Kakao	kanana-2-30B-a3B	1460 ± 30	1
112	OpenBuddy	openbuddy-qwen2.5llamaify-14B-v23.1-200k (Low coverage)	1459 ± 20	7
113	Saltlux	luxia-21.4B-alignment-v1.0 (Low coverage)	1457 ± 19	2
114	RLHFlow	LLaMA3-iterative-DPO-final (Low coverage)	1452 ± 19	2
115	University of Bari	LLaMAntino-3-ANITA-8B-Inst-DPO-ITA (Low coverage)	1450 ± 22	1
116	OpenChat	openchat-3.5-1210	1447 ± 21	5
117	Databricks	DBRX	1446 ± 12	7
117	Korea University	KULLM3 (Low coverage)	1446 ± 19	1
119	CMU	MAmmoTH2-8B-Plus	1441 ± 15	3
120	Bllossom	llama-3-Korean-Bllossom-8B (Low coverage)	1436 ± 18	1
120	MTS AI	multi_verse_model (Low coverage)	1436 ± 19	1
122	Stability AI	StableBeluga2	1434 ± 11	13
123	Nexusflow	Starling-LM-7B-beta	1430 ± 12	2
124	Salesforce	Llama 3 8B SFR Iterative DPO R (Low coverage)	1428 ± 18	2
125	Magpie	Llama 3 8B Magpie Align V0.3 (Low coverage)	1423 ± 18	5
126	UC Berkeley	Starling-LM-7B-alpha	1411 ± 12	1
127	SCB 10X	llama-3-typhoon-v1.5-8B-instruct	1410 ± 20	1
128	Rakuten	RakutenAI-7B-chat	1409 ± 30	1
129	Tsinghua & ByteDance	SALMONN-13B (Low coverage)	1407 ± 21	2
130	Moxoff	Volare (Low coverage)	1405 ± 23	1
131	Intel	neural-chat-7B-v3-3 (Low coverage)	1399 ± 26	2
132	Teknium	OpenHermes-2.5-Mistral-7B	1398 ± 18	3
133	Deci AI	DeciLM-7B-instruct (Low coverage)	1394 ± 29	2
134	Infly	OpenCoder-8B-Instruct (Low coverage)	1393 ± 24	1
135	Baichuan	Baichuan2-13B-Chat	1391 ± 16	3
136	GritLM	GritLM-7B (Low coverage)	1386 ± 18	2
136	KAIST	janus-dpo-7B (Low coverage)	1386 ± 30	4
138	Tilde	TildeOpen-30B	1382 ± 30	1
139	Writer	Palmyra X (43B) (Low coverage)	1369 ± 24	1
140	OpenLLM-France	Lucie-7B	1367 ± 29	1
141	LMSYS	vicuna-33B-v1.3	1363 ± 10	9
142	Open-Orca	Mistral-7B-OpenOrca (Low coverage)	1360 ± 16	1
143	VoiceLab	trurl-2-13B-academic	1339 ± 11	2
144	Kurakura AI	Luth-0.6B-Instruct	1332 ± 30	1
145	NTQ	Nxcode-CQ-7B-orpo (Low coverage)	1326 ± 20	1
146	Mobius Labs	DeepSeek-R1-ReDistill-Llama3-8B-v1.1	1311 ± 26	2
147	H2O.ai	h2o-danube3-4B-chat (Low coverage)	1308 ± 57	4
148	Johns Hopkins University	ALMA-13B-Pretrain (Low coverage)	1306 ± 26	2
149	Avignon Univ	BioMistral-7B	1298 ± 16	1
150	BigCode	starcoder2-15B	1291 ± 18	4
151	Together	Llama 2 7B 32K	1266 ± 12	6
152	RWKV	RWKV-6-World-7B (Low coverage)	1261 ± 45	1
153	Sakana AI	TinySwallow-1.5B (Low coverage)	1238 ± 33	1
154	AI Sweden	gpt-sw3-40B	1231 ± 29	4
155	Singapore Univ	TinyLlama-1.1B-Chat-v1.0	1226 ± 37	2
156	Phind	Phind-CodeLlama-34B-v2	1195 ± 14	1
157	EleutherAI	gpt-neox-20B	1186 ± 10	13
158	BigScience	BLOOMZ-7b1 (Low coverage)	1180 ± 27	5
159	NucleusAI	nucleus-22B-token-500B (Low coverage)	1179 ± 33	1
160	State Space Models	mamba-2.8B-hf	1095 ± 45	1
161	CroissantLLM	CroissantLLMChat-v0.1	1030 ± 31	1
Unranked	AI4Bharat	No ranked model	—	0
Unranked	Boson AI	No ranked model	—	0
Unranked	BSC	No ranked model	—	0
Unranked	Celeris	No ranked model	—	0
Unranked	Cerebras	No ranked model	—	0
Unranked	Cisco	No ranked model	—	0
Unranked	CUHK Shenzhen	No ranked model	—	0
Unranked	DataCanvas	No ranked model	—	0
Unranked	DeepAuto	No ranked model	—	0
Unranked	Du Xiaoman	No ranked model	—	0
Unranked	EPFL	No ranked model	—	0
Unranked	EuroLLM	No ranked model	—	0
Unranked	FuseAI	No ranked model	—	0
Unranked	HUST	No ranked model	—	0
Unranked	John Snow Labs	No ranked model	—	0
Unranked	Krutrim	No ranked model	—	0
Unranked	LAION	No ranked model	—	0
Unranked	Langboat	No ranked model	—	0
Unranked	LLaVA Team	No ranked model	—	0
Unranked	LM Provers	No ranked model	—	0
Unranked	Manus	No ranked model	—	0
Unranked	MedAlpaca	No ranked model	—	0
Unranked	mii-llm	No ranked model	—	0
Unranked	Numina & Hugging Face	No ranked model	—	0
Unranked	NYU	No ranked model	—	0
Unranked	Occiglot	No ranked model	—	0
Unranked	OpenGPT-X	No ranked model	—	0
Unranked	OpenLM Research	No ranked model	—	0
Unranked	OpenThaiGPT	No ranked model	—	0
Unranked	OpenThoughts	No ranked model	—	0
Unranked	Ornith	No ranked model	—	0
Unranked	RedNote	No ranked model	—	0
Unranked	Rhymes AI	No ranked model	—	0
Unranked	Saama AI Labs	No ranked model	—	0
Unranked	Samsung	No ranked model	—	0
Unranked	SenseLLM	No ranked model	—	0
Unranked	SentientAGI	No ranked model	—	0
Unranked	ServiceNow	No ranked model	—	0
Unranked	TensorOpera	No ranked model	—	0
Unranked	The Fin AI	No ranked model	—	0
Unranked	Trillion Labs	No ranked model	—	0
Unranked	UCAS	No ranked model	—	0
Unranked	UCSD	No ranked model	—	0
Unranked	UIUC	No ranked model	—	0
Unranked	USTC	No ranked model	—	0
Unranked	VAIV	No ranked model	—	0
Unranked	Various	No ranked model	—	0
Unranked	Vercel	No ranked model	—	0
Unranked	Vikhr	No ranked model	—	0
Unranked	vikhyatk	No ranked model	—	0
Unranked	VinAI	No ranked model	—	0
Unranked	vivo	No ranked model	—	0
Unranked	WeChat AI	No ranked model	—	0
Unranked	WinGPT	No ranked model	—	0
Unranked	XVERSE	No ranked model	—	0
Unranked	Yandex	No ranked model	—	0
Unranked	Zhejiang University	No ranked model	—	0

Interactive version: theaggregate.ai/companies · How It Works · Data refreshed daily, snapshot 2026-10-05.

Company leaderboard

Sign in to keep your work

Account