发布比较工具

并排比较最多五个模型发布，覆盖每个发布的所有变体，包括智能、价格、输出速度、延迟、上下文窗口等指标。

如需了解我们如何衡量和评估模型，请参阅方法论页面。

比较的发布


	Claude Opus 5.5 (max with fallback)	Claude Opus 5.5 (xhigh with fallback)	Claude Opus 5.5 (high with fallback)	GPT-6 Astra (max)	GPT-6 Astra (xhigh)	Claude Opus 5.5 (medium with fallback)	GPT-6 Astra (high)	GPT-6 Astra (medium)	GPT-6 Astra (low)	Claude Opus 5.5 (low with fallback)
Intelligence Index	58	56	54	53	52	51	51	50	46	42
财务与会计	61	58	56	55	54	54	53	52	49	46
战略与运营	64	62	59	57	57	57	55	54	50	48
法律	63	61	59	59	58	57	58	57	53	52
医疗与健康	55	53	49	53	51	49	51	50	46	43
工程	60	59	56	55	55	54	53	52	48	45
经济学	66	63	61	60	59	60	58	58	55	53
AA-Briefcase v1.1	1801	1761	1683	1568	1543	1622	1505	1460	1255	1276
GDPval-AA v2.1	1866	1829	1704	1574	1557	1585	1525	1487	1391	1234
AutomationBench-AA	69.5%	65.0%	63.2%	68.5%	67.2%	61.2%	66.6%	64.6%	59.1%	52.9%
Terminal-Bench 4.0	59.6%	59.6%	56.6%	59.1%	59.6%	52.5%	54.0%	49.5%	41.9%	31.3%
SciCode	66.9%	65.0%	60.4%	56.5%	55.7%	59.3%	55.4%	54.2%	54.1%	58.6%
Humanity's Last Exam	61.4%	57.5%	55.6%	54.7%	54.6%	54.7%	53.1%	52.7%	49.2%	48.3%
GDP.pdf	26.2%	26.6%	28.8%	31.0%	32.2%	25.6%	31.0%	30.4%	30.4%	25.6%
CritPt	31.7%	31.7%	30.9%	31.7%	31.4%	27.7%	28.9%	29.1%	26.3%	17.7%
AA-Omniscience	46	43	41	43	43	40	44	42	41	39
AA-LCR v1.1	84.7%	84.7%	82.7%	80.7%	80.0%	84.3%	80.0%	79.7%	80.0%	80.7%

每任务成本美元	$5.98	$3.46	$1.82	$3.26	$2.31	$1.34	$1.73	$1.54	$0.82	$0.55
输入价格美元/100 万 Token	$4.00	$4.00	$4.00	$10.00	$10.00	$4.00	$10.00	$10.00	$10.00	$4.00
输出价格美元/100 万 Token	$20.00	$20.00	$20.00	$50.00	$50.00	$20.00	$50.00	$50.00	$50.00	$20.00
缓存命中价格美元/100 万 Token	$0.20	$0.20	$0.20	$1.00	$1.00	$0.20	$1.00	$1.00	$1.00	$0.20
缓存写入价格美元/100 万 Token	$5.00	$5.00	$5.00	$12.50	$12.50	$5.00	$12.50	$12.50	$12.50	$5.00

输出速度Token/秒	96	85	81	58	54	84	56	53	53	81
首 Token 延迟秒	588.68	109.09	24.30	246.99	130.21	10.44	35.20	4.04	2.22	5.38
首个回答 Token 时间秒	588.68	109.09	24.30	246.99	130.21	10.44	35.20	4.04	2.22	5.38
端到端响应时间秒	593.87	115.00	30.50	255.59	139.42	16.41	44.17	13.48	11.58	11.54
每任务耗时秒	795.39	497.53	272.16	473.92	309.45	194.61	210.39	176.76	84.00	80.19

上下文窗口	1M	1M	1M	1M	1M	1M	1M	1M	1M	1M
开发商	Anthropic	Anthropic	Anthropic	OpenAI	OpenAI	Anthropic	OpenAI	OpenAI	OpenAI	Anthropic
许可证	专有	专有	专有	专有	专有	专有	专有	专有	专有	专有
输入模态	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像	支持：文本和图像
输出模态	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本	支持：文本
服务商

智能

Artificial Analysis Intelligence Index · Higher is better

速度

Output tokens per second · Higher is better

每任务成本

Weighted average cost (USD) per Intelligence Index task · Lower is better

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Intelligence Index 与每项任务的成本

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task

Most attractive quadrant

Claude Opus 5.5

GPT-6 Astra

Pareto line

AA-Briefcase v1.1

GDPval-AA v2.1

单任务成本（美元，对数刻度）

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

成本

每项 Intelligence Index 任务的成本

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

速度与延迟

输出速度

Output tokens per second · Higher is better

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API or the median across providers where a first-party API is not available.

能力得分

能力指数

衡量模型在特定能力与行业中的表现

Finance & Accounting Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, Humanity's Last Exam, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Strategy & Ops Index

Incorporates 6 evaluations: GDPval-AA v2.1, AA-Briefcase v1.1, AA-Omniscience, AutomationBench-AA, AA-LCR v1.1, GDP.pdf · Higher is better

Legal Index

Incorporates 7 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, Humanity's Last Exam, AA-LCR v1.1, GDP.pdf, AutomationBench-AA · Higher is better

Healthcare & Medical Index

Incorporates 6 evaluations: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, MLCR-AA, Humanity's Last Exam, AutomationBench-AA · Higher is better

Engineering Index

Incorporates 6 evaluations: AA-Omniscience, Humanity's Last Exam, CritPt, GDPval-AA v2.1, AA-Briefcase v1.1, Terminal-Bench 4.0 · Higher is better

Economics Index

Incorporates 5 evaluations: AA-Omniscience, Humanity's Last Exam, GDPval-AA v2.1, AA-Briefcase v1.1, AA-LCR v1.1 · Higher is better

基准测试

AA-PersonalityBenchNew

模型 Avs模型 B

AA-PersonalityBench：Expressiveness 指数，Claude Opus 5.5 (max with fallback) vs GPT-6 Astra (max)

根据两者的 Elo 差距，每个模型的回答在各特质上被判定为更显著的概率 · 50% 表示势均力敌

Claude Opus 5.5 (max with fallback)

GPT-6 Astra (max)

Lexically Varied

97%

Metaphorical

93%

Eloquent

79%

21%

Literate

59%

41%

Natural Sounding

56%

44%

发布比较工具

智能

速度

每任务成本

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Index 与每项任务的成本

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

成本

每项 Intelligence Index 任务的成本

Cost per Intelligence Index Task

速度与延迟

输出速度

Output Speed

Model Performance Representation

能力得分

能力指数

基准测试

AA-PersonalityBenchNew

AA-PersonalityBench：Expressiveness 指数，Claude Opus 5.5 (max with fallback) vs GPT-6 Astra (max)