Leaderboard

Every model worth knowing, on one table.

Ranked on independent benchmarks first and vendor claims second, next to what each model costs and how fast it answers. Click any model for its full record and sources.

Models tracked: 267
Sourced scores: 2,010
Open weights: 112
Released since July: 81

The table

Sort by what your work needs.

The default ranking uses the Artificial Analysis Intelligence Index, which runs the same set of evaluations on every model with the same harness. Coding teams should sort by SWE-bench Pro or Terminal-Bench. Agent builders should sort by τ²-Bench.

What each benchmark measures →

Showing 185 of 185 language models. v marks a vendor-reported score; everything else is an independent run.

Language model leaderboard, sortable by column
#
1	Claude Opus 5.5Anthropicnew	57.6	90^{v (vendor-reported)}	60	61	—	—	88	—	$4.00 / $20.00	81	1M	2026-09
2	Muse Wonder FullMetanew	56	—	57	—	—	—	—	—	—	—	1M	2026-10
3	Claude Sonnet 5.5Anthropicnew	56	81^{v (vendor-reported)}	71^{v (vendor-reported)}	65^{v (vendor-reported)}	—	—	—	—	$2.00 / $10.00	—	1M	2026-09
4	Claude Fable 5.1Anthropicnew	53.4	81^{v (vendor-reported)}	58	59	94	—	—	1499	$10.00 / $50.00	58	1M	2026-09
5	Gemini 4 ArgonGooglenew	53	—	57	57	—	—	—	1525	$2.00 / $10.00	—	1M	2026-09
6	GPT-6 AstraOpenAInew	52.7	—	59	55	96	—	87	1480	$10.00 / $50.00	58	1.1M	2026-09
7	GPT-6.1 SolOpenAInew	52	—	55	—	—	—	—	1484	$2.00 / $10.00	65	1.1M	2026-09
8	Muse WonderMetanew	52	—	52	—	—	—	—	—	$4.00 / $20.00	83	1M	2026-10
9	Claude Opus 5Anthropic	50.8	79^{v (vendor-reported)}	54	55	93	—	85	1493	$5.00 / $25.00	58	1M	2026-07
10	Claude Fable 5Anthropic	49.6	80^{v (vendor-reported)}	45	56	93	99	—	1506	$10.00 / $50.00	70	1M	2026-06
11	Muse Spark 1.3Metanew	48.1	—	33	49	94	—	—	1493	$1.25 / $4.25	227	1M	2026-09
12	GPT-6 SolOpenAInew	47.5	—	44	48	—	—	83	—	$2.00 / $10.00	87	1.1M	2026-09
13	GPT-5.6 SolOpenAI	47	—	40	50	94	85	83	1483	$4.00 / $20.00	79	1.1M	2026-07
14	Grok 4.7xAInew	46.4	—	26	43	—	—	—	—	$2.00 / $6.00	48	500K	2026-09
15	MiMo-V2.6-ProXiaomiopennew	46.3	—	35	49	—	—	—	—	$0.44 / $0.87	43	1M	2026-09
16	Qwen3.8-MaxAlibabanew	45	68^{v (vendor-reported)}	39	43	93	—	83	1481	$2.00 / $6.00	40	1M	2026-08
17	GLM-5.3Z.aiopennew	45	—	42	42	92	—	—	1483	$1.40 / $4.40	69	1M	2026-08
18	Grok 4.6xAInew	44.3	—	21	43	95	—	—	1456	$2.00 / $6.00	66	500K	2026-08
19	Kimi K3Moonshot AIopen	44	—	13	47	94	—	81	1485	$3.00 / $15.00	35	1M	2026-07
20	Step 5 PreviewStepFunnew	43.7	—	33	47	—	—	76	—	$1.00 / $2.70	77	1M	2026-09
21	Claude Haiku 5.5Anthropicnew	43	—	33	44	—	—	—	—	$0.10 / $0.50	236	1M	2026-10
22	GPT-5.6 TerraOpenAI	42.1	—	35	43	93	86	81	1466	$2.00 / $12.00	92	1.1M	2026-07
23	GLM-5.3-FlashZ.aiopennew	42	—	33	40	91	—	—	1475	$0.15 / $0.50	44	1M	2026-08
24	Claude Opus 4.8Anthropic	41.8	—	24	49	92	94	—	1481	$5.00 / $25.00	56	1M	2026-05
25	Ling-3.1-FlashAnt Groupnew	41	—	—	—	—	—	—	—	—	212	1M	2026-09
26	Gemini 3.8 FlashGooglenew	40.9	—	20	48	95	—	86	1493	$0.75 / $3.75	330	1M	2026-09
27	Qwen3.8-2.4T-A95BAlibabaopennew	40	—	11	42	94	—	—	—	$2.00 / $6.00	40	262K	2026-08
28	Qwen3.8-Flash-NextAlibabaopennew	40	63^{v (vendor-reported)}	25	38	92	—	80	—	self-host	54	262K	2026-08
29	Muse Spark 1.2Metanew	39.6	—	7.1	46	90	—	—	1500	$1.25 / $4.25	236	1M	2026-08
30	Gemini 3.7 FlashGooglenew	39.1	—	14	48	95	—	86	1490	$0.75 / $3.75	339	1M	2026-08
31	DeepSeek V4.1 FlashDeepSeekopennew	39	—	27	39	91^{v (vendor-reported)}	—	77	—	$0.30 / $1.20	235	1M	2026-09
32	Grok 4.5xAI	38.8	—	11	43	93	—	80	1468	$2.00 / $6.00	62	500K	2026-07
33	Claude Sonnet 5Anthropic	38.2	63^{v (vendor-reported)}	14	41	91	—	77	1461	$2.00 / $10.00	70	1M	2026-06
34	Mistral Large 4Mistralnew	38	—	27	35	—	—	—	1429	$1.36 / $4.18	—	1M	2026-10
35	GPT-6 LunaOpenAInew	37.3	—	13	39	—	—	76	—	$0.10 / $0.50	146	1.1M	2026-09
36	GPT-5.6 LunaOpenAI	37.3	—	12	40	91	—	79	1452	$0.20 / $1.20	126	1.1M	2026-07
37	DeepSeek V4 Pro (0813)DeepSeekopennew	36	—	14	41	93	—	—	1463	$1.32 / $3.96	89	1M	2026-08
38	DeepSeek V4 Flash Vision ExpDeepSeekopennew	35	—	12	35	91	—	75	—	self-host	—	1M	2026-08
39	Gemini 3.6 FlashGoogle	34	—	7.1	41	93	—	83	1480	$0.75 / $3.75	212	1M	2026-07
40	DeepSeek V4 Flash 0731DeepSeekopennew	34	—	12	39	91	—	—	1436	self-host	—	1M	2026-07
41	Qwen3.8-27BAlibabaopennew	34	62^{v (vendor-reported)}	5.6	34	91	—	76	1437	$0.50 / $3.00	44	262K	2026-08
42	GLM-5.2Z.aiopen	34	62^{v (vendor-reported)}	1	41	90	99	—	1472	$1.40 / $4.40	77	1M	2026-06
43	Muse Spark 1.1Meta	33.7	—	6.1	46	90	—	—	1493	$1.25 / $4.25	—	1M	2026-07
44	Gemini 3.5 FlashGoogle	32.6	—	6.6	43	92	95	84	1478	$1.50 / $9.00	238	1M	2026-05
45	GPT-5.3 CodexOpenAI	32.5	—	—	43	92	86	79	—	$1.75 / $14.00	120	400K	2026-02
46	K2 Horizon 375B A23BMBZUAIopennew	30.5	43^{v (vendor-reported)}	1.5	32	87	—	—	—	self-host	—	524K	2026-09
47	Gemini 3.1 Pro PreviewGoogle	29.7	54^{v (vendor-reported)}	4	47	94	96	82	1487	$2.00 / $12.00	122	1M	2026-02
48	Qwen3.7-MaxAlibaba	29	61^{v (vendor-reported)}	1.5	41	92	95	—	—	$2.50 / $7.50	208	1M	2026-05
49	MiniMax-M3MiniMaxopen	29	59^{v (vendor-reported)}	2	39	93	89	79	1441	$0.30 / $1.20	179	1M	2026-06
50	GLM-5Z.aiopen	28	—	—	29	82	98	—	1458	$1.00 / $3.20	74	200K	2026-02
51	Inkling-SmallThinking Machines Labopennew	27.8	56^{v (vendor-reported)}	1	33	90	—	74	—	$0.30 / $1.20	202	1M	2026-07
52	JT-4.1 Flash 236B A21BChina Mobile	27.3	—	—	18	85	—	64	—	—	—	256K	2026-07
53	Grok Build 0.1xAI	27.2	—	—	38	90	—	77	—	$1.00 / $2.00	48	256K	2026-06
54	Kimi K2.6Moonshot AIopen	27	—	0.5	38	91	96	79	1460	$0.95 / $4.00	52	262K	2026-04
55	Quasar 438BMultiverse Computingnew	26.7	—	1	19	73	—	—	—	$0.60 / $1.80	170	1M	2026-08
56	AApodex 1.1Apodexnew	26.4	—	0	34	86	—	79	—	$0.30 / $3.00	—	256K	2026-08
57	Kimi K2.7 CodeMoonshot AIopen	26	—	1	35	90	90	—	—	$0.95 / $4.00	64	262K	2026-06
58	MiMo-V2.5-ProXiaomiopen	26	—	0	36	87	94	—	1467	$0.44 / $0.87	47	1M	2026-04
59	Tencent Hy3Tencentopen	25.3	58^{v (vendor-reported)}	0.5	34	90	—	—	1456	$0.08 / $0.33	86	256K	2026-07
60	K2 Horizon MoVA 36B A4BMBZUAIopennew	25.3	—	0	23	82	—	—	—	self-host	—	524K	2026-09
61	MiMo-V2.5Xiaomiopen	25.2	—	0	27	85	91	75	1434	$0.14 / $0.28	50	1M	2026-04
62	Qwen3.7-PlusAlibaba	25	58^{v (vendor-reported)}	1	36	90	93	81	1456	$0.40 / $1.60	70	1M	2026-06
63	InklingThinking Machines Labopen	25	54^{v (vendor-reported)}	1	32	87	—	74	—	$1.00 / $4.05	167	1M	2026-07
64	Grok 4.3xAI	24.9	—	0	37	90	98	78	1443	$1.25 / $2.50	127	1M	2026-04
65	Ling 3.0 FlashAnt Groupopennew	24.9	—	0	24	86	—	—	—	self-host	358	262K	2026-08
66	Ling-3.0-flash-VLAnt Groupopennew	24.6	—	0	22	86	—	79	—	self-host	146	262K	2026-09
67	GPT-5.4 miniOpenAI	24.1	—	2	28	88	83	73	1448	$0.75 / $4.50	175	400K	2026-03
68	Solar Mini 4Upstagenew	24	—	1	—	—	—	—	—	$0.10 / $0.40	208	524K	2026-09
69	MiniMax-M2.7MiniMaxopen	23	—	0	30	87	85	—	1415	$0.30 / $1.20	61	205K	2026-03
70	Nemotron 3 Ultra 550B A55BNVIDIAopen	22.9	—	0.5	28	87	83	—	—	$0.60 / $2.50	140	1M	2026-06
71	Gemini 3.5 Flash-LiteGoogle	22.2	—	1	19	84	—	79	1456	$0.30 / $2.50	350	1M	2026-07
72	KAT-Coder-Pro V2Kuaishou	21.7	—	—	16	86	90	—	—	—	—	256K	2026-03
73	Qwen3.6-27BAlibabaopen	21	54^{v (vendor-reported)}	0	23	84	94	75	—	$0.60 / $3.60	59	262K	2026-04
74	GPT-5.4 nanoOpenAI	20.7	—	0.5	28	82	76	65	1402	$0.20 / $1.25	170	400K	2026-03
75	Step 3.7 FlashStepFunopen	19.5	56^{v (vendor-reported)}	—	21	81	99	75	—	$0.20 / $1.15	197	262K	2026-05
76	LongCat-2.0Meituanopen	19.1	60^{v (vendor-reported)}	0	34	78	—	—	—	$0.75 / $2.95	—	1M	2026-06
77	Gemma 4 31BGoogleopen	19	—	0	24	86	60	73	1451	self-host	35	256K	2026-03
78	Qwen3.6-35B-A3BAlibabaopen	18	—	0	22	84	95	75	—	$0.38 / $2.25	128	262K	2026-04
79	Muse GlimmerMetaopennew	17.5	51^{v (vendor-reported)}	0.5	22	84	—	74	1427	self-host	106	131K	2026-08
80	Claude Haiku 4.5Anthropic	16.9	39	0	10	67	55	59	1415	$1.00 / $5.00	82	200K	2025-10
81	Doubao Seed CodeByteDance	16.9	—	—	14	76	58	68	—	—	—	256K	2025-11
82	Gemma 4 26B A4BGoogleopen	16.7	—	—	19	79	44	69	1438	self-host	—	256K	2026-03
83	Ring-2.6-1TAnt Groupopen	16.6	—	0.5	22	86	92	—	—	self-host	122	262K	2026-05
84	Granite 4.2 30BIBMopennew	14.8	33^{v (vendor-reported)}	—	11	64	—	—	—	$0.16 / $0.65	73	131K	2026-08
85	ERNIE 5.0 Thinking PreviewBaidu	14.3	—	—	13	78	84	65	—	—	—	128K	2025-11
86	Gemma 4 12BGoogleopen	14.2	—	—	16	75	36	70	—	self-host	115	256K	2026-06
87	Mistral Medium 3.5Mistralopen	14.2	—	0	14	75	94	65	—	$1.50 / $7.50	147	256K	2026-04
88	Amazon Nova 2 Pro (Preview)Amazon	14.2	—	—	9.4	79	93	65	—	$1.25 / $10.00	115	256K	2025-11
89	Amazon Nova 2 Omni (Preview)Amazon	13.6	—	—	7	76	80	62	—	$0.30 / $2.50	—	1M	2025-11
90	Amazon Nova 2 LiteAmazon	13.4	—	—	12	81	73	64	—	$0.30 / $2.50	173	1M	2025-10
91	Apriel v1.6 15B ThinkerServiceNowopen	13.4	—	—	11	73	69	—	—	self-host	—	128K	2025-11
92	Command A+Cohereopen	13.1	—	0.5	12	76	81	63	—	self-host	176	128K	2026-05
93	Nemotron 3.5 Lightning 30B A3BNVIDIAopennew	12.9	—	0.5	11	74	—	—	—	$0.07 / $0.22	264	1M	2026-08
94	Nemotron 3 Super 120B A12BNVIDIAopen	12.8	—	0	21	80	68	—	—	$0.30 / $0.90	169	1M	2026-03
95	Mercury 2.5Inceptionnew	12.3	—	0	12	—	—	—	—	$0.25 / $0.75	781	260K	2026-09
96	Seed-OSS-36B-InstructByteDanceopen	12.1	—	—	9.9	73	49	—	—	self-host	41	512K	2025-08
97	Nemotron Cascade 2 30B A3BNVIDIAopen	11.7	—	—	12	76	53	—	—	self-host	—	1M	2026-03
98	gpt-oss-120bOpenAIopen	11.6	—	0	20	78	66	—	1352	self-host	184	131K	2025-08
99	Mistral Small 4Mistralopen	11.3	—	0	9.9	77	41	57	—	$0.15 / $0.60	167	256K	2026-03
100	Granite 4.2 8BIBMopennew	11.1	19^{v (vendor-reported)}	0	9.7	63	—	—	—	$0.06 / $0.25	79	131K	2026-08
101	Trinity Large ThinkingArcee AIopen	10.8	—	0.5	16	75	90	—	—	$0.25 / $0.80	328	512K	2026-04
102	INTELLECT-3Prime Intellectopen	10.6	—	—	13	76	27	—	—	self-host	—	131K	2025-11
103	Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIAopen	10.3	—	—	4.8	47	45	53	—	$0.20 / $1.10	267	256K	2026-04
104	Llama 4 MaverickMetaopen	10	—	0	4.9	67	18	62	1327	self-host	115	1M	2025-04
105	North Mini Code 1.0Cohereopen	9.9	—	0.5	11	76	37	—	—	self-host	90	256K	2026-06
106	DiffusionGemma 26B A4BGoogleopen	9.5	—	—	11	67	—	67	—	self-host	—	256K	2026-06
107	Mistral Large 3Mistralopen	9.3	—	0	4.2	68	25	56	—	$0.50 / $1.50	77	256K	2025-12
108	Granite 4.2 3BIBMopennew	9.1	—	0	6.6	56	—	—	—	$0.03 / $0.12	217	131K	2026-08
109	gpt-oss-20bOpenAIopen	9	—	0	11	69	60	—	1317	self-host	160	131K	2025-08
110	Gemma 4 E4BGoogleopen	8.9	—	—	3.8	58	21	51	—	self-host	50	128K	2026-03
111	Nemotron 3 Nano 30B A3BNVIDIAopen	8.9	—	0	11	76	41	—	—	$0.05 / $0.20	195	1M	2025-12
112	Sarvam 105BSarvam AIopen	8.8	—	—	11	74	47	—	—	self-host	—	128K	2026-03
113	LFM2.5-2.6BLiquid AIopennew	8.4	—	—	6.2	56	—	—	—	self-host	—	128K	2026-08
114	Llama 4 ScoutMetaopen	8.1	—	0	3.8	59	16	53	1321	self-host	100	10M	2025-04
115	Hermes 4 70BNous Researchopen	7.9	—	—	8.8	70	23	—	—	self-host	—	128K	2025-08
116	Gemma 4 E2BGoogleopen	7.8	—	—	4.8	43	21	45	—	self-host	—	128K	2026-03
117	Falcon-H1R-7BTIIopen	7.8	—	—	11	66	28	—	—	self-host	—	256K	2026-01
118	Step3 VL 10BStepFunopen	7.7	—	—	11	69	16	64	—	self-host	—	66K	2026-01
119	ERNIE 4.5 300B A47BBaiduopen	7.5	—	—	3.3	81	0	—	—	self-host	—	131K	2025-06
120	Hermes 4 405BNous Researchopen	7.5	—	—	11	73	22	—	—	$1.00 / $3.00	42	128K	2025-08
121	Nemotron 3 Nano 4BNVIDIAopen	7.4	—	—	4.9	51	28	—	—	self-host	—	262K	2026-03
122	LFM2.5-8B-A1BLiquid AIopen	7.2	—	—	6.9	51	16	—	—	self-host	—	33K	2026-05
123	Olmo 3.1 32B ThinkAi2open	7.1	—	—	6.3	59	0	—	—	self-host	—	66K	2025-12
124	Amazon Nova ProAmazon	7	—	—	3.2	50	14	44	—	$0.80 / $3.20	—	300K	2024-12
125	Command ACohereopen	7	—	—	4	53	15	—	—	$2.50 / $10.00	53	256K	2025-03
126	Amazon Nova LiteAmazon	6.7	—	—	4.3	43	18	38	—	$0.06 / $0.24	168	300K	2024-12
127	Olmo 3.1 32B InstructAi2open	6.5	—	—	5	54	21	—	—	self-host	—	66K	2026-01
128	Phi-4 Mini InstructMicrosoftopen	6.3	—	—	4.5	33	8.2	—	—	self-host	44	128K	—
129	Jamba 1.7 LargeAI21 Labsopen	6.1	—	—	3.7	39	14	—	—	self-host	—	256K	2025-07
130	Ministral 3 14BMistralopen	6	—	0	4.6	57	27	50	—	$0.20 / $0.20	89	256K	2025-12
131	Phi-4Microsoftopen	5.9	—	—	3.8	58	0	—	—	self-host	36	16K	2024-12
132	Amazon Nova MicroAmazon	5.9	—	—	4.6	36	14	—	—	$0.04 / $0.14	265	130K	2024-12
133	LFM2 24B A2BLiquid AIopen	5.9	—	—	4.2	47	11	—	—	self-host	—	33K	2026-02
134	Phi-4 Multimodal InstructMicrosoftopen	5.8	—	—	5	32	—	15	—	self-host	17	128K	2025-02
135	Jamba Reasoning 3BAI21 Labsopen	5.7	—	—	3.8	33	16	—	—	self-host	—	262K	2025-10
136	Olmo 3 7B ThinkAi2open	5.6	—	—	6	52	0	—	—	self-host	—	66K	2025-11
137	Reka Flash 3Reka AIopen	5.6	—	—	4.4	53	0	—	—	$0.20 / $0.80	—	128K	2025-03
138	Ministral 3 8BMistralopen	5.5	—	0	4.3	47	27	46	—	$0.15 / $0.15	97	256K	2025-12
139	Jamba 1.7 MiniAI21 Labsopen	5.2	—	—	4.4	32	13	—	—	self-host	—	258K	2025-07
140	Apertus 70B InstructSwiss AI Initiativeopen	5.1	—	—	5.5	27	13	—	—	$0.82 / $2.92	—	66K	2025-09
141	Molmo 2 8BAi2open	5	—	—	4.3	43	0	38	—	self-host	—	37K	2025-12
142	Ministral 3 3BMistralopen	4.8	—	0	5.4	36	25	38	—	$0.10 / $0.10	209	256K	2025-12
143	Tiny Aya GlobalCohereopen	4.8	—	—	5.2	31	0	—	—	self-host	—	8K	2026-02
144	BeamReflection AInew	—	66^{v (vendor-reported)}	—	36^{v (vendor-reported)}	91^{v (vendor-reported)}	—	—	—	—	—	256K	2026-10
145	Claude Mythos 5.1Anthropicnew	—	—	61^{v (vendor-reported)}	—	—	—	—	—	$10.00 / $50.00	—	1M	2026-09
146	Cogito v2.1 (Reasoning)Deep Cogitoopen	—	—	—	12	77	—	—	—	$1.25 / $1.25	—	128K	2025-11
147	Composer 2.5Cursor	—	—	—	—	—	—	—	—	$0.50 / $2.50	—	200K	—
148	Doubao Seed 2.0 CodeByteDance	—	—	—	—	—	—	—	—	$0.50 / $3.00	—	262K	2026-02
149	Doubao Seed 2.0 LiteByteDance	—	47^{v (vendor-reported)}	—	26^{v (vendor-reported)}	88^{v (vendor-reported)}	—	78^{v (vendor-reported)}	—	$0.25 / $2.00	—	262K	2026-02
150	Doubao Seed 2.0 MiniByteDance	—	—	—	—	—	—	—	—	$0.10 / $0.40	—	262K	2026-02
151	Doubao Seed 2.0 ProByteDance	—	—	—	—	—	—	—	1456	$0.50 / $3.00	—	262K	2026-02
152	Doubao Seed 2.1 LiteByteDance	—	—	—	—	—	—	—	—	—	—	—	—
153	Doubao Seed 2.1 ProByteDance	—	—	—	—	—	—	82^{v (vendor-reported)}	—	—	—	—	2026-06
154	Doubao Seed 2.1 TurboByteDance	—	—	—	—	—	—	80^{v (vendor-reported)}	—	$0.50 / $2.50	—	262K	2026-06
155	Doubao Seed EvolvingByteDance	—	—	—	—	—	—	—	—	—	—	—	—
156	ERNIE 5.0Baidu	—	—	—	—	—	—	—	1446	$0.84 / $3.38	—	128K	—
157	ERNIE 5.1Baidu	—	—	—	—	—	—	—	1468	$0.56 / $2.53	—	—	2026-05
158	FuguSakana AI	—	59^{v (vendor-reported)}	—	47^{v (vendor-reported)}	96^{v (vendor-reported)}	—	—	—	—	—	—	2026-06
159	Fugu MaxSakana AInew	—	—	—	45^{v (vendor-reported)}	96^{v (vendor-reported)}	—	—	—	$2.00 / $6.00	—	—	2026-09
160	Fugu Ultra v2Sakana AInew	—	—	—	56^{v (vendor-reported)}	96^{v (vendor-reported)}	—	—	—	$5.00 / $30.00	—	—	2026-09
161	Gemini 3.1 Flash-LiteGoogle	—	—	—	—	—	—	—	1432	$0.25 / $1.50	—	—	—
162	GLM-4.5-AirZ.aiopen	—	—	—	—	—	—	—	1373	$0.20 / $1.10	—	—	—
163	iFlytek Spark X2.5iFlyteknew	—	—	—	—	—	—	—	—	—	—	256K	2026-09
164	KAT-Coder-Pro V2.5Kuaishou	—	65^{v (vendor-reported)}	—	—	—	—	—	—	$0.74 / $2.96	—	262K	2026-07
165	KAT-Coder-V2.5-DevKuaishouopen	—	46^{v (vendor-reported)}	—	—	—	—	—	—	self-host	—	262K	—
166	Kimi K2.8 PreviewMoonshot AInew	—	—	—	—	—	—	—	—	—	—	1M	2026-09
167	Kolibri-1Aleph Alphaopennew	—	—	—	22^{v (vendor-reported)}	84^{v (vendor-reported)}	95^{v (vendor-reported)}	—	—	self-host	—	1M	2026-10
168	LongCat-2.5-PreviewMeituannew	—	—	—	—	—	—	—	—	$0.75 / $2.95	—	—	2026-09
169	MAI-Code-1.1-FlashMicrosoftnew	—	—	—	—	—	—	—	—	$0.20 / $1.20	—	—	2026-08
170	MAI-Thinking-1Microsoft	—	53^{v (vendor-reported)}	—	—	84^{v (vendor-reported)}	—	—	—	$2.00 / $8.00	—	256K	2026-06
171	MiMo-V2.6-FlashXiaomiopennew	—	—	29^{v (vendor-reported)}	—	—	—	—	—	$0.14 / $0.28	—	1M	2026-09
172	MiMo-V2.6-Pro-UltraSpeedXiaominew	—	—	—	—	—	—	—	—	$4.35 / $8.70	—	1M	2026-09
173	MiniMax M3.1 Flash PreviewMiniMaxnew	—	—	—	—	—	—	—	—	—	—	1M	2026-09
174	Nano Banana 2 (Gemini 3.1 Flash Image)Google	—	—	—	—	—	—	—	—	$0.50 / $60.00	—	—	2026-02
175	Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Google	—	—	—	—	—	—	—	—	$0.25 / $30.00	—	—	2026-06
176	Nano Banana 2.1Googlenew	—	—	—	—	—	—	—	—	$1.50 / $30.00	—	—	2026-10
177	Nano Banana Pro (Gemini 3 Pro Image)Google	—	—	—	—	—	—	—	—	$2.00 / $120.00	—	—	2025-11
178	Qwen3.6-FlashAlibaba	—	—	—	—	—	—	—	—	$0.25 / $1.50	—	1M	2026-04
179	Qwen3.7-FlashAlibaba	—	—	—	—	—	—	—	—	$0.03 / $0.13	—	1M	2026-07
180	Qwen3.8-FlashAlibaba	—	—	—	—	—	—	—	—	$0.15 / $0.47	—	1M	—
181	Qwen3.8-Omni-FlashAlibabanew	—	—	—	—	—	—	—	—	$0.15 / $0.47	—	1M	2026-09
182	Sonar (Agent API, perplexity/sonar)Perplexity	—	—	—	—	—	—	—	—	$0.25 / $2.50	—	—	—
183	Spark-X2.5-4BiFlytekopennew	—	44^{v (vendor-reported)}	—	12^{v (vendor-reported)}	67^{v (vendor-reported)}	75^{v (vendor-reported)}	—	—	self-host	—	1M	2026-09
184	Step 3.5 FlashStepFunopen	—	—	—	—	—	—	—	1394	$0.10 / $0.30	—	262K	2026-02
185	Tencent Hy4 previewTencentopennew	—	66^{v (vendor-reported)}	—	—	92^{v (vendor-reported)}	—	—	—	$0.83 / $2.50	—	1M	2026-08

Price against capability

Most models are not on the frontier.

A model is on the efficient frontier when nothing cheaper scores higher. If a model you use sits well below the line, you are paying for a brand or a feature, not for capability.

Intelligence Index against blended price

92 language models with an independent index score and a published list price. Price is 70% input and 30% output tokens, the planner's default mix.

Closed weights
Open weights
Efficient frontier

View as table

Intelligence Index against blended price
Model	Provider	AA Intelligence Index	Blended $/M	Frontier
Claude Opus 5.5	Anthropic	57.6	$8.80	Yes
Claude Sonnet 5.5	Anthropic	56	$4.40	Yes
Claude Fable 5.1	Anthropic	53.4	$22.00
Gemini 4 Argon	Google	53	$4.40
GPT-6 Astra	OpenAI	52.7	$22.00
GPT-6.1 Sol	OpenAI	52	$4.40
Muse Wonder	Meta	52	$8.80
Claude Opus 5	Anthropic	50.8	$11.00
Claude Fable 5	Anthropic	49.6	$22.00
Muse Spark 1.3	Meta	48.1	$2.15	Yes
GPT-6 Sol	OpenAI	47.5	$4.40
GPT-5.6 Sol	OpenAI	47	$8.80
Grok 4.7	xAI	46.4	$3.20
MiMo-V2.6-Pro	Xiaomi	46.3	$0.57	Yes
Qwen3.8-Max	Alibaba	45	$3.20
GLM-5.3	Z.ai	45	$2.30
Grok 4.6	xAI	44.3	$3.20
Kimi K3	Moonshot AI	44	$6.60
Step 5 Preview	StepFun	43.7	$1.51
Claude Haiku 5.5	Anthropic	43	$0.22	Yes
GPT-5.6 Terra	OpenAI	42.1	$5.00
GLM-5.3-Flash	Z.ai	42	$0.26
Claude Opus 4.8	Anthropic	41.8	$11.00
Gemini 3.8 Flash	Google	40.9	$1.65
Qwen3.8-2.4T-A95B	Alibaba	40	$3.20
Muse Spark 1.2	Meta	39.6	$2.15
Gemini 3.7 Flash	Google	39.1	$1.65
DeepSeek V4.1 Flash	DeepSeek	39	$0.57
Grok 4.5	xAI	38.8	$3.20
Claude Sonnet 5	Anthropic	38.2	$4.40
Mistral Large 4	Mistral	38	$2.21
GPT-6 Luna	OpenAI	37.3	$0.22
GPT-5.6 Luna	OpenAI	37.3	$0.50
DeepSeek V4 Pro (0813)	DeepSeek	36	$2.11
Gemini 3.6 Flash	Google	34	$1.65
Qwen3.8-27B	Alibaba	34	$1.25
GLM-5.2	Z.ai	34	$2.30
Muse Spark 1.1	Meta	33.7	$2.15
Gemini 3.5 Flash	Google	32.6	$3.75
GPT-5.3 Codex	OpenAI	32.5	$5.43
Gemini 3.1 Pro Preview	Google	29.7	$5.00
Qwen3.7-Max	Alibaba	29	$4.00
MiniMax-M3	MiniMax	29	$0.57
GLM-5	Z.ai	28	$1.66
Inkling-Small	Thinking Machines Lab	27.8	$0.57
Grok Build 0.1	xAI	27.2	$1.30
Kimi K2.6	Moonshot AI	27	$1.87
Quasar 438B	Multiverse Computing	26.7	$0.96
Apodex 1.1	Apodex	26.4	$1.11
Kimi K2.7 Code	Moonshot AI	26	$1.87
MiMo-V2.5-Pro	Xiaomi	26	$0.57
Tencent Hy3	Tencent	25.3	$0.16	Yes
MiMo-V2.5	Xiaomi	25.2	$0.18
Qwen3.7-Plus	Alibaba	25	$0.76
Inkling	Thinking Machines Lab	25	$1.92
Grok 4.3	xAI	24.9	$1.63
GPT-5.4 mini	OpenAI	24.1	$1.88
Solar Mini 4	Upstage	24	$0.19
MiniMax-M2.7	MiniMax	23	$0.57
Nemotron 3 Ultra 550B A55B	NVIDIA	22.9	$1.17
Gemini 3.5 Flash-Lite	Google	22.2	$0.96
Qwen3.6-27B	Alibaba	21	$1.50
GPT-5.4 nano	OpenAI	20.7	$0.52
Step 3.7 Flash	StepFun	19.5	$0.49
LongCat-2.0	Meituan	19.1	$1.41
Qwen3.6-35B-A3B	Alibaba	18	$0.94
Claude Haiku 4.5	Anthropic	16.9	$2.20
Granite 4.2 30B	IBM	14.8	$0.31
Mistral Medium 3.5	Mistral	14.2	$3.30
Amazon Nova 2 Pro (Preview)	Amazon	14.2	$3.88
Amazon Nova 2 Omni (Preview)	Amazon	13.6	$0.96
Amazon Nova 2 Lite	Amazon	13.4	$0.96
Nemotron 3.5 Lightning 30B A3B	NVIDIA	12.9	$0.12	Yes
Nemotron 3 Super 120B A12B	NVIDIA	12.8	$0.48
Mercury 2.5	Inception	12.3	$0.40
Mistral Small 4	Mistral	11.3	$0.29
Granite 4.2 8B	IBM	11.1	$0.12
Trinity Large Thinking	Arcee AI	10.8	$0.42
Nemotron 3 Nano Omni 30B A3B Reasoning	NVIDIA	10.3	$0.47
Mistral Large 3	Mistral	9.3	$0.80
Granite 4.2 3B	IBM	9.1	$0.06	Yes
Nemotron 3 Nano 30B A3B	NVIDIA	8.9	$0.10
Hermes 4 405B	Nous Research	7.5	$1.60
Amazon Nova Pro	Amazon	7	$1.52
Command A	Cohere	7	$4.75
Amazon Nova Lite	Amazon	6.7	$0.11
Ministral 3 14B	Mistral	6	$0.20
Amazon Nova Micro	Amazon	5.9	$0.07
Reka Flash 3	Reka AI	5.6	$0.38
Ministral 3 8B	Mistral	5.5	$0.15
Apertus 70B Instruct	Swiss AI Initiative	5.1	$1.45
Ministral 3 3B	Mistral	4.8	$0.10

Independent first

When a lab and an independent evaluator both publish a score, the table shows the independent run. Vendor-only numbers carry a v and usually come from the vendor’s own agent scaffold.

Best published setting

Reasoning models score differently at each effort level. Each cell shows the highest published setting; hover a score to see which setting produced it.

32 benchmarks, one registry

Every score maps to a named benchmark version, so two numbers in the same column were earned on the same test. Older versions stay in their own column.