LLM Leaderboard 2026: Rank & Compare 100+ AI Models Live

Q: How do I compare LLMs?

Sort by Quality Index to find the strongest models overall, then filter by price, speed, or context window to match your constraints. Use the Compare page to put 2–4 finalists head to head.

Model discovery workbench

Move from a crowded market to a defensible shortlist. Rank models by capability, task economics, speed, context, and real provider availability.

Models: 194
Providers: 89
Data snapshot: Sep 14, 2026, 16:24 UTC
Method: Index v4.3
Latest tracked: Sep 11, 2026

Rank by

Show

Model rankings

174 results · sorted by intelligence

Underlined names open model pages. Use the chevron for endpoint details.

#
1	Claude Fable 5.1Anthropic	53.4	81.6	58	$2.37	16s	1.0M	1
2	GPT-6 AstraOpenAI	52.8	76.9	51.5	$3.26	5.5 min	1.0M	1
3	GPT-6 Astra (xhigh)OpenAI	52.5	75.9	50.6	$2.31	2.4 min	1.0M	1
4	GPT-6 Astra (high)OpenAI	51	77.1	48.7	$1.72	44s	1.0M	1
5	Claude Opus 5Anthropic	50.7	78	56.2	$1.10	13s	1.0M	3
6	Claude Fable 5Anthropic	49.7	76.5	51	$8.75	89s	1.0M	3
7	GPT-6 Astra (medium)OpenAI	49.7	76.7	46.3	$1.54	14s	1.0M	1
8	Muse Spark 1.3OpenMeta	48.2	75.8	55.7	$1.60	34s	1.0M	1
9	GPT-5.6 SolOpenAI	47.1	77.4	50.5	$1.99	9.3s	1.0M	3
10	GPT-6 Astra (low)OpenAI	46	75.7	39.5	$0.818	12s	1.0M	1
11	Muse Spark 1.3 (xhigh)OpenMeta	45.2	76.5	51.8	$1.37	32s	1.0M	1
12	GLM-5.3OpenZ AI	44.9	74.8	53.4	$2.01	41s	1.0M	15
13	Grok 4.6 (high)SpaceXAI	44.4	76.8	53.4	$1.86	38s	500K	1
14	Grok 4.6 (xhigh)SpaceXAI	44.3	75.9	52.7	$2.32	47s	500K	1
15	GPT-5.6 Sol (xhigh)OpenAI	44.1	78.3	47.8	$1.18	49s	1.0M	2
16	Kimi K3OpenKimi	43.8	76.2	50.6	$2.00	69s	1.0M	11
17	Grok 4.6 (medium)SpaceXAI	43	74.4	51.2	$1.50	33s	500K	1
18	GPT-5.6 Sol (high)OpenAI	42.5	77.2	44.7	$0.808	24s	1.0M	2
19	GPT-5.6 TerraOpenAI	42.3	76.7	43.7	$1.40	6.8s	1.0M	2
20	GLM-5.3-FlashOpenZ AI	41.9	71.5	51.2	$0.253	27s	1.0M	18
21	Gemini 3.8 Flash (high)Google	41.2	76.3	41.1	$1.24	14s	1.0M	1
22	Qwen3.8 MaxOpenAlibaba	40.3	71.8	49.6	$2.67	69s	1.0M	1
23	Qwen3.8 2.4T A95BOpenAlibaba	40	71.9	50.4	$2.16	68s	984K	6
24	Qwen3.8-Flash-NextOpenAlibaba	39.9	73.1	53.9	$0.372	51s	1.0M	2
25	Muse Spark 1.2 (xhigh)OpenMeta	39.8	72.2	44	$0.975	28s	1.0M	1
26	Gemini 3.7 Flash (medium)Google	39.6	71.5	N/A	N/A	6.2s	1.0M	1
27	GPT-5.6 Sol (medium)OpenAI	39.5	76.3	40.5	$0.505	14s	1.0M	1
28	DeepSeek V4.1 FlashOpenDeepSeek	39.5	N/A	N/A	$0.265	13s	1.0M	2
29	Gemini 3.7 Flash (high)Google	39.4	76.1	36.4	$0.925	12s	1.0M	1
30	Grok 4.5 (high)SpaceXAI	39.1	72.4	42.1	$1.04	20s	500K	1
31	GLM-5.2OpenZ AI	38.6	68.8	39.4	N/A	5.6s	1.0M	18
32	Claude Sonnet 5Anthropic	38.4	71.5	44.3	$0.509	9.8s	1.0M	4
33	GPT-5.6 Terra (xhigh)OpenAI	38.2	70.6	42.2	$0.632	29s	1.0M	2
34	GPT-5.6 LunaOpenAI	37.5	71.4	42.7	$0.178	5.7s	1.0M	2
35	Gemini 3.7 Flash (low)Google	36.9	71	N/A	N/A	2.6s	1.0M	1
36	DeepSeek V4 Pro 0813OpenDeepSeek	36.3	68.8	42.3	$0.674	36s	1.0M	10
37	Agnes 3.0 FlashSapiens AI	35.5	N/A	N/A	N/A	12s	1.0M	1
38	Grok 4.6 (low)SpaceXAI	35.4	66.3	42.2	$0.475	15s	500K	1
39	DeepSeek V4 Flash VisionOpenDeepSeek	35	65	48.1	$0.314	13s	1.0M	2
40	GPT-5.6 Luna (xhigh)OpenAI	34.8	68.6	39.5	$0.085	52s	1.0M	2
41	Kimi K3 (low)OpenKimi	34.5	72	30.7	N/A	69s	1.0M	3
42	GPT-5.6 Terra (high)OpenAI	34.5	67.1	37.6	$0.338	9.5s	1.0M	2
43	DeepSeek V4 Flash 0731OpenDeepSeek	34.5	69.1	41.7	$0.220	13s	1.0M	15
44	Gemini 3.6 Flash (high)Google	34.3	69.2	30.2	$0.929	19s	1.0M	1
45	Qwen3.8 27B (xhigh)OpenAlibaba	33.9	68.1	46.5	$0.820	67s	1.0M	5
46	GPT-5.6 Sol (low)OpenAI	33.8	69.7	32.6	$0.261	12s	1.0M	1
47	Gemini 3.5 Flash (medium)Google	33.6	N/A	N/A	N/A	15s	1.0M	1
48	GPT-5.3 Codex (xhigh)OpenAI	32.5	N/A	N/A	N/A	54s	400K	1
49	GPT-5.6 Luna (high)OpenAI	32.4	63.3	35.6	$0.044	12s	1.0M	2
50	DeepSeek V4 ProOpenDeepSeek	30.9	59.4	27.7	$0.121	9.1s	1.0M	8
51	GPT-5.6 Terra (medium)OpenAI	30.4	64.7	31.4	$0.183	7.7s	1.0M	1
52	Gemini 3.1 Pro PreviewGoogle	30.4	68.8	10.3	$0.675	30s	1.0M	2
53	MiniMax-M3OpenMiniMax	29.6	58.6	30.8	$0.508	30s	1.0M	13
54	Solar Pro 4Upstage	28.2	52.7	N/A	N/A	43s	512K	1
55	GPT-5.6 Terra (low)OpenAI	27.9	58.1	27.3	$0.144	7.7s	1.0M	1
56	Qwen3.8 27B (medium)OpenAlibaba	27.8	56.1	45.3	$0.903	58s	1.0M	2
57	Agnes 2.5 Pro AlphaOpenSapiens AI	27.8	58.8	29.4	N/A	18s	1.0M	1
58	Quasar 438BOpenMultiverse Computing	27.1	61.2	32.7	$2.02	15s	1.0M	1
59	GPT-5.5 InstantOpenAI	26.8	39.4	19.2	$0.692	20s	400K	1
60	Qwen3.8 27B (low)OpenAlibaba	26.5	58.2	39.9	$0.834	60s	1.0M	2
61	MiMo-V2.5-ProOpenXiaomi	26.4	60.2	22.7	$0.054	21s	1.0M	3
62	Kimi K2.7 CodeOpenKimi	26.3	60.8	22.5	$0.541	60s	262K	9
63	Inkling SmallOpenThinking Machines	26.1	52.9	25	$0.089	19s	524K	3
64	Qwen3.7 PlusOpenAlibaba	25.8	55.9	19.7	$0.325	38s	1.0M	1
65	Hy3OpenTencent	25.8	58.8	25.6	$0.072	32s	262K	4
66	Inkling (xhigh)OpenThinking Machines	25.5	52.1	24.3	$0.607	36s	1.0M	5
67	GPT-5.6 Luna (medium)OpenAI	25.5	50.7	25.4	$0.016	7.8s	1.0M	1
68	Ling 3.0 FlashOpenInclusionAI	24.9	50.6	21	N/A	11s	262K	2
69	Ling-3.0-flash-VLOpenInclusionAI	24.8	57	30	N/A	20s	262K	1
70	Grok 4.3 (medium)SpaceXAI	24.8	N/A	N/A	N/A	16s	1.0M	3
71	Grok 4.3 (low)SpaceXAI	24.3	N/A	N/A	N/A	9.8s	1.0M	3
72	Gemini 3.5 FlashGoogle	23.8	N/A	N/A	N/A	3.3s	1.0M	1
73	Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA	23.4	49.3	21.7	$0.585	20s	262K	4
74	Claude Sonnet 4.6Anthropic	23.3	N/A	N/A	N/A	14s	1.0M	1
75	Gemini 3.5 Flash-LiteGoogle	22.7	49.3	15.9	$0.123	8.7s	1.0M	1
76	Qwen3.8 27BOpenAlibaba	22.4	44.6	N/A	N/A	15s	1.0M	2
77	MiMo-V2.5OpenXiaomi	22.3	56.8	17.4	$0.019	62s	1.0M	3
78	Qwen3.6 27B (Reasoning)OpenAlibaba	21.9	53.7	20.1	$0.621	1.9 min	262K	5
79	GPT-5.6 Luna (low)OpenAI	21.5	44.2	17.9	$0.0098	6.3s	1.0M	1
80	Qwen3.5 397B A17BOpenAlibaba	21.4	N/A	N/A	N/A	8.3s	262K	6
81	Qwen3.5 Omni PlusOpenAlibaba	20.4	N/A	N/A	N/A	7.8s	256K	1
82	o3OpenAI	20.2	N/A	N/A	N/A	10s	200K	2
83	Qwen3.6 27BOpenAlibaba	19.8	46.6	N/A	N/A	13s	262K	5
84	LongCat 2.0OpenLongCat	19.7	45.3	15.9	$0.059	N/A	262K	1
85	Step 3.7 FlashOpenStepFun	19.5	39.6	N/A	N/A	29s	256K	2
86	Qwen3.5 397B A17B (Reasoning)OpenAlibaba	19.1	48.2	10.6	$0.475	49s	262K	7
87	DeepSeek V4 FlashOpenDeepSeek	18.9	N/A	N/A	N/A	N/A	1.0M	2
88	Qwen3.6 35B A3B (Reasoning)OpenAlibaba	18.8	41.9	15	$0.475	48s	262K	6
89	Muse Glimmer (high)OpenMeta	18.1	49	10.5	$0.055	28s	131K	3
90	Qwen3.5 122B A10BOpenAlibaba	17.7	43.3	N/A	N/A	5.9s	262K	2
91	Claude 4.5 Haiku (Reasoning)Anthropic	17.6	43.9	10.3	$0.208	24s	200K	4
92	Ring-2.6-1TOpenInclusionAI	17.3	42.8	12.9	$0.287	24s	262K	1
93	Gemma 4 26B A4B (Reasoning)OpenGoogle	16.7	39.3	N/A	N/A	N/A	1.0M	5
94	Qwen3.5 122B A10B (Reasoning)OpenAlibaba	16.2	45.7	9.6	$0.325	22s	262K	4
95	Claude 4.5 HaikuAnthropic	15.4	N/A	N/A	N/A	7.2s	200K	4
96	Gemma 4 31B (Reasoning)OpenGoogle	15.4	43.4	6.7	$0.0000	64s	262K	11
97	Qwen3.6 35B A3BOpenAlibaba	15.2	28.1	N/A	N/A	5.7s	262K	5
98	Qwen3.5 35B A3BOpenAlibaba	15.1	37	N/A	N/A	5.2s	262K	3
99	Mistral Medium 3.5OpenMistral	14.9	46.9	9.4	$0.437	19s	262K	1
100	Granite 4.2 30BOpenIBM	14.8	29.9	N/A	N/A	33s	131K	1
101	Grok 4.3SpaceXAI	14.5	35.2	15.8	$0.137	5.2s	1.0M	2
102	Nova 2.0 Pro Preview (medium)Amazon	14.2	34	N/A	N/A	34s	256K	1
103	Gemma 4 12B (Reasoning)OpenGoogle	14.2	31	N/A	N/A	26s	262K	1
104	Command A+OpenCohere	13.9	27.8	3.6	$0.0000	10s	200K	1
105	Gemma 4 31BOpenGoogle	13.9	33.2	N/A	N/A	12s	262K	10
106	Qwen3.5 9B (Reasoning)OpenAlibaba	13.7	28.7	N/A	N/A	29s	262K	2
107	Nemotron 3.5 LightningOpenNVIDIA	13.6	26.8	6.1	$0.093	9.5s	1.0M	6
108	Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA	13.6	37.7	4.1	$1.06	19s	262K	2
109	Nova 2.0 Lite (high)Amazon	13.4	23	N/A	N/A	34s	1.0M	1
110	Qwen3.5 9BOpenAlibaba	13.3	23.5	N/A	N/A	6.0s	262K	1
111	Qwen3.5 4B (Reasoning)OpenAlibaba	13.1	22.6	N/A	N/A	1.6 min	262K	1
112	Gemma 4 26B A4BOpenGoogle	13.1	N/A	N/A	N/A	8.0s	1.0M	6
113	North Mini CodeOpenCohere	12.8	36.5	1.1	N/A	31s	256K	1
114	Nova 2.0 Pro Preview (low)Amazon	12.8	25.9	N/A	N/A	31s	256K	1
115	Qwen3.5 Omni FlashOpenAlibaba	12.5	N/A	N/A	N/A	4.1s	256K	1
116	Nova 2.0 Lite (medium)Amazon	12.5	N/A	N/A	N/A	32s	1.0M	1
117	gpt-oss-120b (high)OpenOpenAI	12.3	30.4	6.2	$0.107	14s	131K	15
118	Ling 3.0 TinyOpenInclusionAI	11.9	26.5	7.1	$0.0000	18s	262K	1
119	Nova 2.0 Lite (low)Amazon	11.8	N/A	N/A	N/A	27s	1.0M	1
120	Granite 4.2 8BOpenIBM	11.8	22.4	3.7	$0.024	30s	131K	2

1Claude Fable 5.1Anthropicintelligence53.4Task $2.37Response 16s1.0M context1 provider

2GPT-6 AstraOpenAIintelligence52.8Task $3.26Response 5.5 min1.0M context1 provider

3GPT-6 Astra (xhigh)OpenAIintelligence52.5Task $2.31Response 2.4 min1.0M context1 provider

4GPT-6 Astra (high)OpenAIintelligence51Task $1.72Response 44s1.0M context1 provider

5Claude Opus 5Anthropicintelligence50.7Task $1.10Response 13s1.0M context3 providers

6Claude Fable 5Anthropicintelligence49.7Task $8.75Response 89s1.0M context3 providers

7GPT-6 Astra (medium)OpenAIintelligence49.7Task $1.54Response 14s1.0M context1 provider

8Muse Spark 1.3OpenMetaintelligence48.2Task $1.60Response 34s1.0M context1 provider

9GPT-5.6 SolOpenAIintelligence47.1Task $1.99Response 9.3s1.0M context3 providers

10GPT-6 Astra (low)OpenAIintelligence46Task $0.818Response 12s1.0M context1 provider

11Muse Spark 1.3 (xhigh)OpenMetaintelligence45.2Task $1.37Response 32s1.0M context1 provider

12GLM-5.3OpenZ AIintelligence44.9Task $2.01Response 41s1.0M context15 providers

13Grok 4.6 (high)SpaceXAIintelligence44.4Task $1.86Response 38s500K context1 provider

14Grok 4.6 (xhigh)SpaceXAIintelligence44.3Task $2.32Response 47s500K context1 provider

15GPT-5.6 Sol (xhigh)OpenAIintelligence44.1Task $1.18Response 49s1.0M context2 providers

16Kimi K3OpenKimiintelligence43.8Task $2.00Response 69s1.0M context11 providers

17Grok 4.6 (medium)SpaceXAIintelligence43Task $1.50Response 33s500K context1 provider

18GPT-5.6 Sol (high)OpenAIintelligence42.5Task $0.808Response 24s1.0M context2 providers

19GPT-5.6 TerraOpenAIintelligence42.3Task $1.40Response 6.8s1.0M context2 providers

20GLM-5.3-FlashOpenZ AIintelligence41.9Task $0.253Response 27s1.0M context18 providers

21Gemini 3.8 Flash (high)Googleintelligence41.2Task $1.24Response 14s1.0M context1 provider

22Qwen3.8 MaxOpenAlibabaintelligence40.3Task $2.67Response 69s1.0M context1 provider

23Qwen3.8 2.4T A95BOpenAlibabaintelligence40Task $2.16Response 68s984K context6 providers

24Qwen3.8-Flash-NextOpenAlibabaintelligence39.9Task $0.372Response 51s1.0M context2 providers

25Muse Spark 1.2 (xhigh)OpenMetaintelligence39.8Task $0.975Response 28s1.0M context1 provider

26Gemini 3.7 Flash (medium)Googleintelligence39.6Task N/AResponse 6.2s1.0M context1 provider

27GPT-5.6 Sol (medium)OpenAIintelligence39.5Task $0.505Response 14s1.0M context1 provider

28DeepSeek V4.1 FlashOpenDeepSeekintelligence39.5Task $0.265Response 13s1.0M context2 providers

29Gemini 3.7 Flash (high)Googleintelligence39.4Task $0.925Response 12s1.0M context1 provider

30Grok 4.5 (high)SpaceXAIintelligence39.1Task $1.04Response 20s500K context1 provider

Showing 120 of 174 models.

Reading the field

How to use this LLM ranking page

Start with capability, then force the economics

Frontier models often cluster tightly on raw intelligence. Cost per task, response time, provider coverage, and context length usually create the real shortlist.

Move from discovery to a defensible final pick

Use Explore to find the shape of the market, then move into Compare or use the Agentic Fit Finder when task-level reliability and cost matter more than chat quality alone.

Field notes

Questions builders ask

What is the best LLM in 2026?

Claude Fable 5.1 leads the overall quality ranking right now. The best model for you depends on your use case — coding, cost, speed, and context length all shift the answer.

How do I compare LLMs?

Sort by Quality Index for overall strength, then filter by price, speed, or context window to match your constraints. Move to the Compare page to put 2–4 finalists head to head.

Which AI model has the best quality-to-price ratio?

Open-weight models like DeepSeek, Qwen, and Llama often lead on quality-to-price. For agentic workflows, cost per task is usually a better filter than token price alone.

How often is this leaderboard updated?

Data is pulled from Artificial Analysis and refreshed automatically. New models appear as soon as they have benchmark scores and provider endpoints.

Ranking library

Current live rankings

Focused rankings for the decisions engineers actually make.

01CodingBest LLM for CodingLiveCodeBench, Terminal-Bench, SciCode, and engineering fit.Open ranking 02OpenBest Open Source LLMOpen-weight models for self-hosting and lower-cost deployment.Open ranking 03LocalBest Local LLMModels mapped to Mac, RTX, workstation, and server hardware.Open ranking 04Local codingBest Local LLM for CodingCoding strength filtered through realistic hardware limits.Open ranking 05OllamaBest Ollama ModelsLow-friction local models for coding, reasoning, and chat.Open ranking 06ContextLargest Context Window LLMLong documents, codebases, and retrieval-heavy work.Open ranking 07AgentsBest Agentic ModelsTool use, multi-step execution, and autonomous workflows.Open ranking