Claude Haiku 5.5 vs. GPT-5.6 Terra：发布比较

Claude Haiku 5.5 与 GPT-5.6 Terra 发布比较：分别包含 5 个和 6 个模型，各自具有不同的智能、性能与价格特征。下方对全部 11 个模型的关键指标进行了比较。

智能方面，Claude Haiku 5.5 得分最高：Claude Haiku 5.5 (Max, Default Fallback)（43），GPT-5.6 Terra 为 GPT-5.6 Terra (Max)（42）。
输出速度方面，Claude Haiku 5.5 最快：Claude Haiku 5.5 (Max, Default Fallback)（243 token/秒），GPT-5.6 Terra 为 GPT-5.6 Terra (Max)（108 token/秒）。
单任务成本方面，Claude Haiku 5.5 最低：Claude Haiku 5.5 (Low, Default Fallback)（$0.02），GPT-5.6 Terra 为 GPT-5.6 Terra (Non-reasoning)（$0.14）。

比较的发布


	Claude Haiku 5.5 (max with fallback)	GPT-5.6 Terra (max)	Claude Haiku 5.5 (xhigh with fallback)	GPT-5.6 Terra (xhigh)	Claude Haiku 5.5 (high with fallback)	Claude Haiku 5.5 (medium with fallback)	GPT-5.6 Terra (high)	GPT-5.6 Terra (medium)	Claude Haiku 5.5 (low with fallback)	GPT-5.6 Terra (low)	GPT-5.6 Terra (non-reasoning)
Intelligence Index	43	42	41	38	38	34	34	30	29	27	21
财务与会计	44	41	41	40	38	35	37	32	29	30	21
战略与运营	40	47	39	44	37	34	40	34	28	31	23
法律	42	41	40	40	38	36	37	34	31	31	22
医疗与健康	--	40	--	38	--	--	34	31	--	27	--
工程	45	43	43	38	39	35	34	31	31	28	21
经济学	51	45	49	44	46	43	42	37	37	34	24
AA-Briefcase v1.1	1578	1333	1533	1334	1443	1372	1202	1039	1112	1013	947
GDPval-AA v2.1	1620	1453	1511	1449	1420	1277	1379	1269	1125	1110	1100
AutomationBench-AA	35.4%	59.6%	36.0%	47.1%	33.7%	28.6%	42.0%	33.5%	22.9%	29.1%	12.5%
Terminal-Bench 4.0	32.8%	35.4%	29.3%	10.1%	21.7%	15.2%	1.5%	1.0%	12.6%	1.5%	0.5%
SciCode	55.0%	55.0%	51.7%	52.3%	48.7%	49.0%	52.4%	50.5%	49.2%	49.9%	45.1%
Humanity's Last Exam	44.4%	42.9%	42.7%	41.9%	37.3%	33.8%	38.5%	33.3%	27.0%	29.2%	11.4%
GDP.pdf	20.8%	24.0%	18.2%	24.6%	17.2%	15.2%	20.8%	17.0%	11.2%	17.6%	11.2%
CritPt	18.9%	30.0%	22.6%	27.1%	18.6%	12.9%	22.9%	17.4%	9.1%	9.4%	2.0%
AA-Omniscience	11	0	6	−3	6	4	−3	−5	3	−7	−23
AA-LCR v1.1	82.7%	83.0%	78.3%	79.0%	77.3%	77.3%	77.7%	74.0%	72.0%	71.3%	58.7%

每任务成本美元	$0.21	$1.40	$0.12	$0.63	$0.08	$0.05	$0.34	$0.18	$0.02	$0.14	$0.14
输入价格美元/100 万 Token	$0.10	$2.00	$0.10	$2.00	$0.10	$0.10	$2.00	$2.00	$0.10	$2.00	$2.00
输出价格美元/100 万 Token	$0.50	$12.00	$0.50	$12.00	$0.50	$0.50	$12.00	$12.00	$0.50	$12.00	$12.00
缓存命中价格美元/100 万 Token	$0.01	$0.20	$0.01	$0.20	$0.01	$0.01	$0.20	$0.20	$0.01	$0.20	$0.20
缓存写入价格美元/100 万 Token	$0.125	$2.50	$0.125	$2.50	$0.125	$0.125	$2.50	$2.50	$0.125	$2.50	$2.50

输出速度Token/秒	243	108	188	102	137	137	91	92	181	89	92
首 Token 延迟秒	415.30	132.62	87.15	22.37	26.15	12.63	2.93	1.76	6.28	1.55	0.76
首个回答 Token 时间秒	415.30	132.62	87.15	22.37	26.15	12.63	2.93	1.76	6.28	1.55	0.76
端到端响应时间秒	417.35	137.24	89.80	27.28	29.81	16.29	8.46	7.18	9.04	7.17	6.20
每任务耗时秒	427.18	352.33	291.49	203.30	235.33	144.82	123.52	59.24	61.83	43.42	33.90

上下文窗口	1M	1M	1M	1M	1M	1M	1M	1M	1M	1M	1M
开发商	Anthropic	OpenAI	Anthropic	OpenAI	Anthropic	Anthropic	OpenAI	OpenAI	Anthropic	OpenAI	OpenAI
许可证	专有	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	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	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 Haiku 5.5

GPT-5.6 Terra

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 Artificial Analysis 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

财务与会计指数

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

战略与运营指数

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

法律指数

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

医疗与健康指数

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

工程指数

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

经济学指数

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

Claude Haiku 5.5 vs. GPT-5.6 Terra：发布比较

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