Claude Sonnet 5.5 vs. Gemma 4 26B A4B：发布比较

Claude Sonnet 5.5 与 Gemma 4 26B A4B 发布比较：分别包含 5 个和 2 个模型，各自具有不同的智能、性能与价格特征。下方对全部 7 个模型的关键指标进行了比较。

智能方面，Claude Sonnet 5.5 得分最高：Claude Sonnet 5.5 (Max, Default Fallback)（56），Gemma 4 26B A4B 为 Gemma 4 26B A4B (Reasoning)（17）。
输出速度方面，Claude Sonnet 5.5 最快：Claude Sonnet 5.5 (Max, Default Fallback)（140 token/秒），Gemma 4 26B A4B 为 Gemma 4 26B A4B (Non-reasoning)（98 token/秒）。

比较的发布


	Claude Sonnet 5.5 (max with fallback)	Claude Sonnet 5.5 (xhigh with fallback)	Claude Sonnet 5.5 (high with fallback)	Claude Sonnet 5.5 (medium with fallback)	Claude Sonnet 5.5 (low with fallback)	Gemma 4 26B A4B	Gemma 4 26B A4B (non-reasoning)
Intelligence Index	56	52	47	41	36	17*	13*
财务与会计	57	52	47	41	38	--	--
战略与运营	60	56	52	45	41	--	--
法律	56	52	49	44	41	--	--
医疗与健康	58	--	--	--	--	--	--
工程	58	54	48	42	38	--	--
经济学	61	56	52	47	44	19	--
AA-Briefcase v1.1	1817	1751	1636	1441	1272	315	--
GDPval-AA v2.1	1838	1728	1549	1323	1175	566	--
AutomationBench-AA	71.8%	65.5%	59.4%	54.9%	49.4%	1.9%	--
Terminal-Bench 4.0	63.6%	57.1%	43.9%	29.8%	20.7%	--	--
SciCode	61.0%	57.3%	53.7%	52.9%	49.1%	40.0%	--
Humanity's Last Exam	55.0%	50.0%	45.8%	39.8%	36.2%	19.3%	11.5%
GDP.pdf	25.8%	24.6%	25.2%	20.2%	16.0%	--	--
CritPt	31.4%	31.1%	24.6%	16.9%	11.4%	0.0%	0.0%
AA-Omniscience	32	24	21	20	19	−51	−62
AA-LCR v1.1	82.7%	79.7%	78.0%	76.3%	76.0%	65.7%	42.3%

每任务成本美元	$5.46	$2.01	$0.88	$0.48	$0.35	--	--
输入价格美元/100 万 Token	$2.00	$2.00	$2.00	$2.00	$2.00	$0.10	$0.13
输出价格美元/100 万 Token	$10.00	$10.00	$10.00	$10.00	$10.00	$0.37	$0.40
缓存命中价格美元/100 万 Token	$0.10	$0.10	$0.10	$0.10	$0.10	$0.05	$0.07
缓存写入价格美元/100 万 Token	$2.50	$2.50	$2.50	$2.50	$2.50	--	--

输出速度Token/秒	140	113	111	109	110	67	98
首 Token 延迟秒	380.34	35.54	11.62	1.11	0.96	0.98	1.04
首个回答 Token 时间秒	380.34	35.54	11.62	1.11	0.96	30.63	1.04
端到端响应时间秒	383.92	39.95	16.11	5.71	5.52	38.04	6.13
每任务耗时秒	888.62	426.53	216.42	129.14	85.55	--	--

上下文窗口	1M	1M	1M	1M	1M	256k	256k
开发商	Anthropic	Anthropic	Anthropic	Anthropic	Anthropic	Google	Google
许可证	专有	专有	专有	专有	专有	开放	开放
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、图像和视频	支持：文本、图像和视频
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
服务商						+4	+5

* 估计值

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

Estimate (independent evaluation forthcoming)

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 Sonnet 5.5

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

Claude Sonnet 5.5 vs. Gemma 4 26B A4B：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