Artificial Analysis LLM Leaderboard

Shown as Artificial Analysis publishes it — their metrics, their models. A model name opens its page here; the arrow opens the original board.

Published by

Artificial Analysis

Licence

not stated

Updates

continuous

Fetched

2026-10-02

Entries

688 · 1 unmatched

Filter modelsOrder byArtificial Analysis Intelligence IndexGPQA DiamondHumanity's Last ExamSciCodeCritPtτ³-BankingTerminal-Bench 2.1GDPval-AAAA-LCRAA-Omniscience IndexTerminal-Bench 4.0AA-Omniscience Non-Hallucination RateIFBenchMMMU-ProBlended Price, USD per 1M tokens (3:1 input:output)Output SpeedTime To First TokenMatched models only

679 scored · 9 not tested

#	Model	Artificial Analysis Intelligence Index ↑	GPQA Diamond ↑	Humanity's Last Exam ↑	SciCode ↑	CritPt ↑	τ³-Banking ↑	Terminal-Bench 2.1 ↑	GDPval-AA ↑	AA-LCR ↑	AA-Omniscience Index ↑	Terminal-Bench 4.0 ↑	AA-Omniscience Non-Hallucination Rate ↑	IFBench ↑	MMMU-Pro ↑	Blended Price, USD per 1M tokens (3:1 input:output) ↓	Output Speed ↑	Time To First Token ↓
1	Claude Opus 5.5↗Default · adaptive · maxAnthropic	57.6	—	61.4%	66.9%	31.7%	—	—	67.3%	84.7%	46.4	59.6%	41.4%	—	87.7%	$8.00	93/s	679.56s
2	Claude Sonnet 5.5↗Default · adaptive · maxAnthropic	56.0	—	55.0%	61.0%	31.4%	—	—	67.0%	82.7%	32.3	63.6%	53.0%	—	—	$4.00	139/s	441.14s
3	Claude Opus 5.5↗Default · adaptive · xhighAnthropic	56.0	—	57.5%	65.0%	31.7%	—	—	66.0%	84.7%	42.6	59.6%	34.3%	—	86.6%	$8.00	79/s	137.34s
4	Claude Opus 5.5↗Default · adaptive · highAnthropic	53.6	—	55.6%	60.4%	30.9%	—	—	59.6%	82.7%	40.6	56.6%	32.4%	—	85.8%	$8.00	74/s	37.37s
5	Claude Fable 5.1↗Default · adaptive · maxAnthropic	53.4	93.7%	59.1%	63.1%	29.7%	47.2%	91.4%	61.7%	85.3%	43.5	52.0%	27.4%	—	—	$20.00	68/s	214.52s
6	Claude Fable 5.1↗Default · adaptive · xhighAnthropic	53.2	93.4%	58.7%	60.9%	31.1%	45.8%	91.0%	61.0%	83.0%	42.4	55.1%	29.5%	—	—	$20.00	66/s	111.92s
7	GPT 6 Astra↗maxOpenAI	52.7	96.1%	54.7%	56.5%	31.7%	41.4%	88.4%	52.1%	80.7%	43.4	59.1%	48.7%	—	86.9%	$20.00	54/s	341.88s
8	Gemini 4 Argon↗highGoogle	52.6	—	57.1%	61.8%	27.1%	—	—	55.6%	79.7%	42.4	57.1%	84.9%	—	—	$4.00	—	—
9	GPT 6 Astra↗xhighOpenAI	52.4	96.3%	54.6%	55.7%	31.4%	43.1%	89.1%	50.8%	80.0%	43.4	59.6%	51.7%	—	86.2%	$20.00	49/s	181.00s
10	Claude Sonnet 5.5↗Default · adaptive · xhighAnthropic	51.9	—	50.0%	57.3%	31.1%	—	—	61.4%	79.7%	23.5	57.1%	37.1%	—	—	$4.00	105/s	32.46s
11	GPT 6.1 Sol↗maxOpenAI	51.8	—	52.9%	54.2%	31.7%	—	—	53.8%	83.0%	41.5	56.1%	45.7%	—	86.0%	$4.00	63/s	291.08s
12	Claude Opus 5.5↗Default · adaptive · mediumAnthropic	51.2	—	54.7%	59.3%	27.7%	—	—	53.8%	84.3%	40.3	52.5%	31.6%	—	85.7%	$8.00	72/s	22.18s
13	Claude Fable 5.1↗Default · adaptive · highAnthropic	51.2	90.6%	55.9%	58.7%	30.3%	43.1%	89.9%	55.9%	83.7%	40.8	52.0%	31.2%	—	—	$20.00	54/s	27.13s
14	GPT 6.1 Sol↗xhighOpenAI	51.0	—	52.6%	55.7%	31.7%	—	—	50.5%	79.7%	40.9	54.0%	49.1%	—	85.1%	$4.00	53/s	110.28s
15	GPT 6 Astra↗highOpenAI	50.9	94.9%	53.1%	55.4%	28.9%	40.0%	89.9%	49.2%	80.0%	43.7	54.0%	55.2%	—	86.4%	$20.00	45/s	58.39s
16	Claude Opus 5↗adaptive · maxAnthropic	50.8	93.2%	54.9%	56.4%	29.1%	42.1%	89.1%	60.4%	79.3%	37.1	49.0%	39.2%	—	84.7%	$10.00	58/s	68.35s
17	GPT 6.1 Sol↗highOpenAI	50.2	—	51.4%	55.8%	30.0%	—	—	49.3%	82.3%	41.5	51.5%	50.6%	—	84.9%	$4.00	58/s	57.61s
18	Claude Opus 5↗adaptive · xhighAnthropic	49.7	93.7%	54.4%	55.7%	27.7%	43.3%	88.0%	58.8%	80.3%	35.4	46.5%	40.5%	—	84.0%	$10.00	56/s	44.82s
19	Claude Fable 5↗Opus 4.8 · adaptive · maxAnthropic	49.6	92.6%	55.5%	61.0%	28.6%	38.1%	84.6%	54.8%	82.3%	43.3	42.4%	36.4%	63.5%	—	$20.00	66/s	97.24s
20	GPT 6 Astra↗mediumOpenAI	49.6	93.9%	52.7%	54.2%	29.1%	35.5%	89.5%	48.4%	79.7%	42.2	49.5%	53.5%	—	85.1%	$20.00	45/s	6.07s
21	Claude Fable 5.1↗Default · adaptive · mediumAnthropic	48.9	88.6%	53.8%	56.4%	29.1%	41.0%	88.0%	51.8%	84.7%	37.6	44.9%	30.9%	—	—	$20.00	54/s	9.11s
22	Claude Opus 5↗adaptive · highAnthropic	48.1	93.7%	52.8%	55.4%	28.3%	44.7%	87.6%	54.0%	79.0%	33.7	46.0%	38.8%	—	82.4%	$10.00	54/s	19.82s
23	Muse Spark 1.3↗maxMeta	48.1	93.5%	48.7%	58.8%	24.9%	50.5%	84.3%	58.6%	83.0%	25.0	33.3%	67.1%	—	—	$2.00	152/s	51.47s
24	GPT 6.1 Sol↗mediumOpenAI	47.8	—	49.9%	53.2%	27.7%	—	—	46.6%	83.3%	40.0	48.0%	48.4%	—	83.9%	$4.00	53/s	5.50s
25	GPT 6 Sol↗maxOpenAI	47.6	—	47.9%	57.6%	30.9%	—	—	50.3%	83.7%	27.1	43.9%	39.9%	—	82.9%	$4.00	93/s	140.78s
26	GPT 5.6 Sol↗maxOpenAI	47.0	94.1%	49.5%	57.1%	32.3%	44.3%	88.0%	54.4%	84.0%	22.0	39.9%	7.8%	72.7%	83.4%	$8.00	80/s	112.99s
27	Claude Fable 5.1↗Default · adaptive · lowAnthropic	46.8	88.1%	48.9%	56.7%	27.7%	39.0%	85.0%	47.5%	82.3%	34.1	40.4%	34.4%	—	—	$20.00	53/s	4.71s
28	Claude Sonnet 5.5↗Default · adaptive · highAnthropic	46.8	—	45.8%	53.7%	24.6%	—	—	52.6%	78.0%	20.9	43.9%	35.4%	—	—	$4.00	102/s	12.20s
29	Grok 4.7↗xhighSpaceXAI	46.4	—	43.1%	57.4%	17.7%	—	—	59.8%	76.7%	32.0	25.8%	70.7%	—	—	$3.00	81/s	49.32s
30	Grok 4.7↗highSpaceXAI	46.3	—	42.3%	57.8%	18.0%	—	—	60.5%	77.0%	30.9	24.7%	67.6%	—	—	$3.00	78/s	33.12s
31	MiMo V2.6 Pro↗Xiaomi	46.3	—	49.4%	60.9%	26.6%	—	—	59.3%	86.3%	8.4	34.8%	59.4%	—	—	$0.54	46/s	4.08s
32	GPT 6 Astra↗lowOpenAI	45.8	93.1%	49.2%	54.1%	26.3%	32.0%	88.0%	43.3%	80.0%	40.5	41.9%	53.1%	—	84.6%	$20.00	46/s	2.82s
33	Qwen 3.8 Max↗0902Alibaba	45.4	92.8%	43.1%	52.1%	17.7%	47.8%	88.8%	58.6%	80.3%	12.0	38.9%	71.2%	—	82.8%	$3.00	39/s	2.69s
34	Muse Spark 1.3↗xhighMeta	45.1	94.1%	47.5%	59.7%	26.0%	47.2%	85.4%	55.9%	83.0%	23.1	16.7%	68.5%	—	82.0%	$2.00	140/s	39.81s
35	Claude Opus 5↗adaptive · mediumAnthropic	44.8	91.9%	51.3%	51.5%	26.9%	38.6%	86.1%	48.8%	82.0%	31.0	34.3%	39.3%	—	81.6%	$10.00	52/s	5.27s
36	GLM 5.3↗maxZ AI	44.8	91.7%	42.3%	59.0%	19.1%	50.3%	83.9%	57.8%	79.7%	14.3	41.9%	70.4%	—	—	$2.15	71/s	3.42s
37	Grok 4.6↗highSpaceXAI	44.3	94.9%	42.9%	56.5%	17.1%	50.7%	88.4%	55.5%	80.3%	30.5	21.2%	65.7%	—	—	$3.00	77/s	29.97s
38	GPT 6 Sol↗xhighOpenAI	44.2	—	46.3%	55.1%	28.0%	—	—	47.8%	81.3%	26.7	30.3%	41.1%	—	82.5%	$4.00	90/s	50.76s
39	Grok 4.6↗xhighSpaceXAI	44.2	93.5%	44.1%	53.0%	19.7%	43.3%	88.0%	56.6%	81.0%	29.3	17.2%	76.0%	—	—	$3.00	72/s	30.32s
40	GPT 5.6 Sol↗xhighOpenAI	44.0	93.1%	47.3%	57.1%	28.6%	38.1%	89.5%	52.4%	82.3%	21.0	24.7%	8.1%	71.0%	82.7%	$8.00	78/s	50.35s
41	Step 5↗previewStepFun	43.7	—	46.5%	58.9%	20.9%	—	—	53.3%	88.3%	16.4	33.3%	57.0%	—	76.4%	$1.43	86/s	3.07s
42	Kimi K3↗maxKimi	43.6	93.5%	46.9%	59.5%	23.4%	46.0%	85.0%	52.0%	88.7%	19.7	12.6%	46.8%	—	80.5%	$6.00	34/s	4.53s
43	Grok 4.6↗mediumSpaceXAI	42.8	93.5%	42.1%	55.9%	17.7%	44.3%	84.3%	55.2%	81.0%	28.0	13.1%	76.0%	—	—	$3.00	76/s	24.09s
44	GPT 6 Sol↗highOpenAI	42.4	—	44.1%	54.9%	25.4%	—	—	44.8%	83.7%	26.8	26.3%	41.9%	—	82.0%	$4.00	87/s	17.74s
45	GPT 5.6 Sol↗highOpenAI	42.3	92.8%	46.0%	57.8%	25.7%	36.7%	87.3%	49.0%	81.7%	20.4	20.7%	8.8%	69.2%	81.8%	$8.00	76/s	13.29s
46	Claude Opus 5.5↗Default · adaptive · lowAnthropic	42.3	—	48.3%	58.6%	17.7%	—	—	36.2%	80.7%	38.9	31.3%	32.4%	—	84.7%	$8.00	76/s	12.39s
47	Grok 4.7↗lowSpaceXAI	42.2	—	39.2%	54.9%	13.4%	—	—	53.7%	80.0%	29.6	16.2%	60.1%	—	78.2%	$3.00	78/s	8.92s
48	GPT 6.1 Sol↗lowOpenAI	42.1	—	47.4%	53.2%	24.9%	—	—	39.9%	84.0%	37.6	30.8%	48.4%	—	83.1%	$4.00	60/s	2.61s
49	GPT 5.6 Terra↗maxOpenAI	42.1	92.5%	42.9%	55.0%	30.0%	40.2%	88.0%	47.7%	83.0%	0.1	35.4%	12.1%	71.2%	80.7%	$4.50	100/s	176.80s
50	GLM 5.3 Flash↗Z AI	41.8	91.2%	39.9%	51.6%	15.4%	47.2%	84.3%	57.3%	80.0%	7.5	32.8%	72.4%	—	—	$0.24	54/s	3.21s
51	Claude Opus 4.8↗adaptive · maxAnthropic	41.8	92.0%	48.7%	54.4%	20.9%	34.2%	84.6%	46.9%	77.7%	28.8	21.7%	60.7%	62.2%	—	$10.00	61/s	63.28s
52	Gemini 3.8 Flash↗highGoogle	40.9	95.3%	47.8%	56.6%	18.3%	44.9%	87.6%	45.6%	81.3%	29.6	19.7%	44.8%	—	85.6%	$1.50	249/s	18.95s
53	Claude Sonnet 5.5↗Default · adaptive · mediumAnthropic	40.8	—	39.8%	52.9%	16.9%	—	—	41.2%	76.3%	20.1	29.8%	48.8%	—	—	$4.00	99/s	1.23s
54	Claude Opus 4.7↗adaptive · maxAnthropic	40.7est	91.4%	42.3%	—	12.0%	34.6%	83.1%	41.9%	78.7%	27.3	—	57.7%	58.6%	78.8%	$10.00	47/s	25.34s
55	Qwen 3.8 Max↗Alibaba	40.2	92.7%	43.0%	53.2%	20.0%	51.3%	81.3%	55.4%	78.3%	3.4	18.7%	58.3%	—	82.3%	$3.00	39/s	2.53s
56	Qwen 3.8 Max↗Alibaba	39.9	93.5%	42.4%	54.1%	20.0%	49.1%	82.0%	54.9%	80.3%	4.3	11.1%	60.8%	—	—	$3.00	40/s	2.81s
57	Qwen 3.8 Flash Next↗Alibaba	39.8	92.3%	38.0%	50.6%	11.1%	45.4%	86.1%	55.6%	79.7%	-9.7	25.3%	54.7%	—	79.8%	$0.23	54/s	2.46s
58	GPT 6 Sol↗mediumOpenAI	39.8	—	41.0%	53.8%	24.6%	—	—	42.4%	82.3%	27.0	18.7%	43.2%	—	80.4%	$4.00	—	—
59	Gemini 3.8 Flash↗mediumGoogle	39.8	93.5%	42.1%	55.1%	12.3%	45.8%	83.9%	45.4%	84.0%	28.6	19.7%	48.1%	—	84.2%	$1.50	—	—
60	Gemini 3.7 Flash↗mediumGoogle	39.6est	92.1%	39.0%	59.8%	9.4%	35.5%	78.3%	42.0%	83.0%	23.7	—	34.1%	—	84.7%	$1.50	283/s	5.62s
61	Muse Spark 1.2↗xhighMeta	39.6	90.4%	45.5%	57.4%	17.7%	34.8%	80.1%	49.1%	79.0%	27.2	7.1%	66.7%	—	—	$2.00	242/s	14.10s
62	DeepSeek V4.1 Flash↗maxDeepSeek	39.5	—	39.2%	51.9%	14.3%	—	—	55.0%	84.0%	-5.3	26.8%	3.5%	—	77.0%	$0.52	209/s	0.95s
63	Claude Opus 5↗adaptive · lowAnthropic	39.4	88.9%	43.4%	49.2%	23.1%	30.3%	76.4%	39.7%	81.3%	28.6	26.3%	37.8%	—	79.8%	$10.00	55/s	2.87s
64	GPT 5.6 Sol↗mediumOpenAI	39.2	92.6%	42.2%	57.4%	22.9%	36.5%	86.1%	45.1%	80.3%	19.4	14.6%	9.2%	69.6%	81.4%	$8.00	73/s	5.46s
65	Gemini 3.7 Flash↗highGoogle	39.1	94.5%	47.9%	57.2%	14.3%	32.8%	85.8%	43.6%	81.7%	26.5	13.6%	35.5%	—	85.5%	$1.50	303/s	11.67s
66	GPT 5.4↗xhighOpenAI	39.0est	92.0%	43.7%	—	23.4%	39.6%	78.3%	36.6%	82.0%	5.8	—	8.3%	73.9%	78.4%	$5.63	89/s	192.94s
67	Grok 4.5↗highSpaceXAI	38.8	93.1%	42.7%	55.0%	15.4%	42.1%	81.6%	43.5%	79.3%	25.3	10.6%	45.9%	—	80.4%	$3.00	55/s	9.83s
68	GPT 5.5↗xhighOpenAI	38.4	93.5%	45.8%	55.8%	27.1%	39.0%	84.3%	41.8%	84.3%	20.5	14.6%	11.0%	75.9%	79.9%	$11.25	85/s	52.31s
69	Claude Sonnet 5↗adaptive · maxAnthropic	38.2	91.1%	41.3%	54.3%	16.9%	37.3%	80.5%	47.5%	82.0%	16.4	14.1%	60.6%	—	77.3%	$4.00	86/s	173.83s
70	GPT 6 Luna↗maxOpenAI	38.1	—	38.5%	54.6%	19.4%	—	—	46.9%	83.3%	0.7	12.6%	23.3%	—	79.7%	$0.20	131/s	128.78s
71	GPT 5.6 Terra↗xhighOpenAI	38.0	90.8%	41.9%	52.3%	27.1%	29.7%	80.1%	46.4%	79.0%	-3.0	10.1%	11.0%	66.3%	79.5%	$4.50	91/s	28.50s
72	MiMo V2.6 Flash↗Xiaomi	37.9	—	35.1%	51.3%	12.0%	—	—	55.0%	74.3%	-12.7	22.7%	45.6%	—	73.1%	$0.18	51/s	3.70s
73	GPT 5.6 Luna↗maxOpenAI	37.3	91.1%	39.5%	53.6%	20.6%	31.1%	80.9%	48.2%	83.7%	-10.3	11.6%	7.4%	—	78.6%	$0.45	125/s	107.35s
74	GPT 5.5↗highOpenAI	37.0	93.2%	45.0%	56.1%	25.4%	36.7%	79.4%	41.5%	84.3%	18.8	9.1%	10.9%	71.6%	81.1%	$11.25	87/s	21.23s
75	Gemini 3.7 Flash↗lowGoogle	36.9est	90.1%	35.1%	55.7%	5.7%	29.5%	79.8%	40.3%	78.7%	22.1	—	32.3%	—	84.9%	$1.50	290/s	1.18s
76	DeepSeek V4 Pro↗0813 · maxDeepSeek	36.0	92.8%	41.0%	51.0%	18.0%	39.6%	78.7%	47.1%	80.3%	0.8	14.1%	5.2%	—	—	$1.98	107/s	1.75s
77	Claude Sonnet 5.5↗Default · adaptive · lowAnthropic	35.9	—	36.2%	49.1%	11.4%	—	—	33.9%	76.0%	19.4	20.7%	49.8%	—	—	$4.00	91/s	1.31s
78	Grok 4.6↗lowSpaceXAI	35.1	87.9%	27.6%	49.4%	5.7%	38.1%	75.3%	45.1%	80.7%	25.9	3.0%	69.4%	—	—	$3.00	68/s	7.44s
79	DeepSeek V4 Flash Vision↗maxDeepSeek	34.8	91.3%	34.5%	49.7%	10.9%	41.0%	74.2%	51.7%	81.3%	-17.6	12.1%	8.5%	—	74.8%	$0.66	215/s	0.93s
80	GPT 6 Luna↗xhighOpenAI	34.6	—	34.3%	51.7%	17.4%	—	—	43.3%	80.0%	-1.8	8.1%	17.6%	—	79.7%	$0.20	138/s	22.85s
81	GPT 5.6 Luna↗xhighOpenAI	34.6	89.5%	37.0%	50.5%	20.6%	28.7%	77.9%	44.2%	81.7%	-10.8	3.5%	7.5%	—	78.6%	$0.45	113/s	53.97s
82	Claude Sonnet 5↗adaptive · xhighAnthropic	34.4	—	39.0%	54.1%	15.4%	—	—	42.2%	76.7%	3.2	7.1%	41.4%	—	—	$4.00	73/s	25.45s
83	DeepSeek V4 Flash↗0731 · maxDeepSeek	34.3	90.8%	38.6%	50.3%	16.6%	39.4%	78.7%	46.3%	79.7%	-14.3	12.1%	8.3%	—	—	$0.66	216/s	0.97s
84	GLM 5.3↗lowZ AI	34.3	—	36.6%	42.0%	14.6%	—	—	40.2%	72.7%	-8.4	34.8%	36.0%	—	—	$2.15	67/s	3.13s
85	GPT 5.6 Terra↗highOpenAI	34.2	89.6%	38.5%	52.4%	22.9%	28.7%	75.7%	43.1%	77.7%	-3.5	1.5%	10.2%	64.4%	79.1%	$4.50	85/s	3.47s
86	GPT 6 Sol↗lowOpenAI	34.2	—	34.9%	50.2%	16.3%	—	—	35.3%	79.3%	26.5	9.1%	49.3%	—	80.2%	$4.00	78/s	1.91s
87	Gemini 3.6 Flash↗highGoogle	34.0	92.8%	40.8%	53.4%	10.6%	29.9%	77.5%	38.2%	80.0%	22.1	7.1%	44.4%	—	83.2%	$1.50	189/s	18.39s
88	JT 4.1 Flash↗236B A21B · onChina Mobile	33.9	—	43.6%	52.0%	14.0%	—	—	40.8%	78.3%	20.6	3.0%	43.0%	—	—	—	—	—
89	GPT 5.5↗mediumOpenAI	33.8	92.6%	42.4%	54.5%	18.6%	29.9%	80.5%	35.9%	83.0%	18.1	5.1%	10.4%	71.0%	81.2%	$11.25	83/s	7.03s
90	Muse Spark 1.1↗xhighMeta	33.7	89.8%	46.2%	58.8%	15.1%	31.8%	77.9%	35.4%	77.7%	28.1	6.1%	50.0%	—	—	$2.00	124/s	30.18s
91	GLM 5.2↗maxZ AI	33.7	89.5%	41.1%	51.2%	20.9%	34.6%	77.9%	42.9%	78.3%	4.4	1.0%	73.7%	73.3%	—	$2.15	84/s	3.10s
92	Qwen 3.8↗27B · xhighAlibaba	33.7	90.5%	33.9%	46.6%	5.4%	48.0%	79.8%	46.2%	82.0%	-10.0	5.6%	69.7%	—	76.3%	$1.13	46/s	3.92s
93	Gemini 3.5 Flash↗mediumGoogle	33.6est	92.1%	41.3%	—	10.9%	—	—	—	74.3%	20.8	—	38.2%	74.6%	83.9%	$3.38	211/s	14.17s
94	Motif 3↗Motif Technologies	33.6est	83.4%	37.0%	42.2%	6.6%	35.3%	74.9%	30.9%	76.0%	10.2	—	71.6%	—	—	—	—	—
95	GPT 5.6 Sol↗lowOpenAI	33.5	89.8%	39.4%	56.4%	14.9%	29.1%	76.8%	39.5%	78.0%	18.9	1.0%	10.6%	66.5%	81.0%	$8.00	72/s	2.72s
96	Gemini 3.8 Flash↗lowGoogle	33.5	92.0%	37.1%	55.0%	4.0%	33.2%	83.1%	39.7%	80.7%	21.4	10.1%	35.4%	—	84.5%	$1.50	—	—
97	GPT 6 Luna↗highOpenAI	32.9	—	32.9%	50.3%	15.4%	—	—	42.3%	79.3%	-5.5	4.5%	15.6%	—	77.5%	$0.20	135/s	12.42s
98	Gemini 3.5 Flash↗highGoogle	32.6	92.2%	42.7%	53.9%	13.1%	32.2%	78.7%	34.2%	73.3%	21.2	6.6%	37.8%	76.3%	84.3%	$3.38	216/s	16.19s
99	GPT 5.3 Codex↗xhighOpenAI	32.5est	91.5%	42.5%	—	16.9%	—	—	—	83.3%	10.9	—	10.8%	75.4%	78.5%	$4.81	95/s	68.50s
100	Motif 3↗betaMotif Technologies	32.3est	86.9%	40.4%	—	6.0%	—	70.8%	29.3%	74.7%	-17.8	—	50.9%	—	—	—	—	—

Show all 688 rows

What each column measures

Artificial Analysis Intelligence Index

Higher is better

GPQA Diamond

Higher is better · Reasoning and knowledge

Humanity's Last Exam

Higher is better · Reasoning and knowledge

SciCode

Higher is better · Coding

CritPt

Higher is better · Reasoning and knowledge

τ³-Banking

Higher is better · Agentic work

Terminal-Bench 2.1

Higher is better · Agentic coding

GDPval-AA

Higher is better · Agentic work

AA-LCR

Higher is better · Long context

AA-Omniscience Index

Higher is better · Factuality

Terminal-Bench 4.0

Higher is better

AA-Omniscience Non-Hallucination Rate

Higher is better · Factuality

IFBench

Higher is better · Instruction following

MMMU-Pro

Higher is better · Multimodal

Blended Price, USD per 1M tokens (3:1 input:output)

Lower is better · Cost

Output Speed

Higher is better · Speed

Time To First Token

Lower is better · Speed