AI Model Leaderboards

Rankings across capability, popularity, adoption, economic footprint, speed, and value. Updated every 6 hours.

Use this page when you want the best models for a specific goal, not just the biggest names.

Active Only Include Non-Active

Capability

Pure technical ability across benchmarks, arenas, and agent tasks.

Popularity

Blended public attention plus real-world traction and trend persistence.

Adoption

Observed practical footprint across providers, routing, and sustained usage proxies.

Economic Footprint

Evidence-weighted economic presence combining adoption, monetization, and distribution.

Value

Capability relative to cost for practical buyers comparing utility per dollar.

Ranking signals refreshed43m agoJun 17, 2026pipeline sync

Start with a lens above, keep Active Only on for the cleanest public view, and open deeper tables only when you want the supporting details.

Structured benchmark-backed rows

Signal-backed rows without full normalized tables

Benchmark-expected rows still catching up

Benchmark badges show how directly comparable a row is. Structured means normalized benchmark rows, Provider-reported means official benchmark claims without a normalized table yet, Arena only means competitive signal without broad benchmark coverage, and Pending means the model is tracked but benchmark ingestion is still catching up.

All Models LLMs Image Gen Vision

49 models

#▲	Model	Category	Economic Footprint	Popularity	Adoption	Value	Cat. Rank	Agent	Downloads
2	GPT-4 Turbo OpenAI	Large Language Models	67.8	65.8	68.0	22.9	#48	32.8	—
5	GPT-4o OpenAI	Multimodal	63.8	62.6	68.0	39.4	#18	38.4	—
6	GPT-4.5 OpenAIArchived	Large Language Models	76.6	47.2	68.7	7.7	#51	34.5	0
11	o3 OpenAI	Large Language Models	57.0	59.4	64.0	43.8	#30	62.9	—
12	GPT-4.1 OpenAI	Large Language Models	57.0	61.7	64.0	43.4	#27	56.9	—
25	GPT-5.5 OpenAI	Large Language Models	54.3	53.2	59.0	32.9	#1	63.8	—
26	Claude 3.5 Sonnet AnthropicArchived	Multimodal	69.7	55.1	70.9	37.3	#35	43.7	0
26	Command A CohereOpen	Large Language Models	54.3	55.6	54.9	40.8	#26	21.8	—
32	Command R+ CohereOpen	Large Language Models	53.6	57.6	69.7	—	#171	—	189K
45	CodeLlama 70B MetaOpen	Code	51.0	65.4	80.0	—	#8	33.2	410.0M
46	GPT-5.1 OpenAI	Large Language Models	51.0	50.7	61.2	48.3	#14	58.9	—
47	llama-3-70b-instruct MetaOpen	Large Language Models	51.0	72.3	80.0	—	#99	55.8	40.0M
51	GPT-5.2 OpenAI	Large Language Models	50.6	54.1	60.9	44.9	#5	66.4	—
58	GPT-5 OpenAI	Large Language Models	50.3	49.1	62.5	46.8	#64	62.0	—
60	llama-3.1-405b-instruct MetaOpen	Large Language Models	50.3	64.4	79.5	—	#103	30.0	15.0M
63	Phi 4 MicrosoftOpen	Large Language Models	49.7	63.0	59.6	84.9	#20	8.3	—
67	Gemma 2 9B GoogleOpen	Large Language Models	49.4	60.3	81.4	—	#97	41.1	4.1M
68	Grok 4.1 Fast xAI	Multimodal	49.1	36.6	54.6	67.3	#9	43.1	—
73	GPT-5.4 OpenAI	Large Language Models	48.8	50.8	59.7	40.3	#2	71.3	—
88	Llama 3.3 70B MetaOpen	Large Language Models	48.0	67.1	77.7	—	#95	31.9	3.2M
91	Qwen2.5-72B AlibabaOpen	Large Language Models	47.8	66.8	75.1	—	#101	54.9	2.8M
93	DeepSeek-V3 DeepSeekOpen	Large Language Models	47.7	74.4	75.4	—	#39	41.8	320.0M
95	DeepSeek-R1 DeepSeekOpen	Large Language Models	47.3	63.9	75.1	—	#58	56.0	4.5M
98	Gemma 3 27B GoogleOpen	Large Language Models	46.9	61.4	77.9	—	#79	23.7	560.0M
100	Llama 4 Scout MetaOpen	Multimodal	46.8	55.3	61.1	78.4	#14	17.0	—
114	Llama 4 Maverick MetaOpen	Large Language Models	45.9	71.7	76.1	—	#54	18.3	2100.0M
116	Kimi K2 Moonshot AI	Large Language Models	45.7	50.5	52.6	57.8	#37	35.3	—
121	Qwen3 Max QwenOpen	Large Language Models	45.1	50.2	55.3	50.8	#42	59.5	—
126	DeepSeek V3.2 Speciale DeepSeekOpen	Large Language Models	44.7	49.2	55.9	68.2	#10	52.8	—
136	Qwen3-235B AlibabaOpen	Large Language Models	44.0	62.3	72.3	—	#56	55.1	180.0M
149	grok-4 xAI	Large Language Models	43.1	41.8	62.1	—	#121	36.3	61K
153	Qwen3 Max Thinking QwenOpen	Large Language Models	42.6	48.1	53.5	50.6	#38	51.6	—
167	Kimi K2.6 Moonshot AIOpen	Multimodal	41.7	49.0	49.5	58.4	#1	68.0	—
175	MiniMax M2 MiniMaxOpen	Large Language Models	41.1	53.8	51.9	66.2	#21	45.8	—
178	Qwen3.7 Plus QwenOpen	Multimodal	40.6	47.0	51.9	66.8	#4	43.7	—
180	Qwen3.7 Max QwenOpen	Large Language Models	40.5	47.0	52.1	46.1	#29	46.7	—
193	gemma-4-31B-it GoogleOpen	Multimodal	39.8	55.0	73.1	79.2	#19	—	7.7M
206	MiniMax M2.5 MiniMaxOpen	Large Language Models	39.3	50.3	50.4	72.7	#19	49.6	—
218	MiniMax M1 MiniMaxOpen	Large Language Models	38.8	42.6	53.6	60.9	#82	—	—
222	MiniMax M2.1 MiniMax	Large Language Models	38.6	51.1	51.1	63.5	#12	41.8	—
228	Qwen3.6 Plus QwenOpen	Multimodal	38.2	47.3	52.8	62.4	#8	39.4	—
230	deepseek-v3.1 DeepSeekOpen	Large Language Models	38.1	46.9	66.3	—	#167	34.6	488K
234	MiniMax M3 MiniMax	Multimodal	37.7	49.3	49.0	67.2	#2	31.9	—
252	qwen3-235b-a22b-instruct-2507 QwenOpen	Large Language Models	37.0	47.5	65.9	—	#153	33.1	1.1M
254	MiniMax M2.7 MiniMax	Large Language Models	36.9	49.8	50.0	64.7	#15	41.3	—
256	QwQ 32B QwenOpen	Large Language Models	36.9	57.3	53.0	—	#61	57.6	—
284	GLM 5.2 Z.ai	Large Language Models	34.6	34.3	45.3	43.6	#11	54.2	—
285	DeepSeek-V3.2 DeepSeekOpen	Large Language Models	34.6	50.9	64.2	—	#173	45.5	869K
318	DeepSeek-V4-Pro DeepSeekOpen	Large Language Models	31.8	52.2	63.9	—	#128	64.3	2.8M