65 个模型

Google 模型：智能、性能与价格

Name: Artificial Analysis Intelligence Index
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

Google

智能

Gemini 3.8 Flash (high)

Gemini 3.8 Flash (medium)

Gemini 3.8 Flash (low)

Gemini 3.1 Pro Preview

Gemini 3.5 Flash-Lite

Artificial Analysis Intelligence Index

速度

Gemini 3.5 Flash-Lite

337 Token/秒

Gemini 3.8 Flash (high)

285 Token/秒

Gemma 4 12B (Non-reasoning)

124 Token/秒

Gemini 3.1 Pro Preview

116 Token/秒

Gemma 4 12B

114 Token/秒

输出速度

成本

Gemini 3.5 Flash-Lite

$0.12

Gemini 3.1 Pro Preview

$0.67

Gemini 3.8 Flash (medium)

$0.93

Gemini 3.8 Flash (high)

$1.24

Gemma 4 12B (Non-reasoning)

每项 Intelligence Index 任务

Artificial Analysis 已对 Google 的 65 个模型进行基准测试。下方对这些模型的关键指标进行了比较。

智能方面，Google 得分最高的模型是 Gemini 3.8 Flash (high)（41）。
输出速度方面，最快的模型是 Gemini 3.5 Flash-Lite（337 token/秒）。
延迟方面，Gemma 4 E4B (Non-reasoning)（0.80 秒）的首个答案 token 等待时间最短。
价格方面，Gemini 3.5 Flash-Lite（$0.12）的单任务成本最低。不同模型的价格差异最高可达 10 倍。

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 vs. Cost per Intelligence Index Task

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

Most attractive quadrant

Pareto line

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

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better

速度与延迟

Output Speed

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: AA-Omniscience, GDPval-AA v2.1, AA-Briefcase v1.1, 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

Google 的所有发布

更多详情

							权重		服务商基准测试
Gemini 3.8 Flash (high)	Google	41	-	1M	US$0.6	285	不可用		查看
Gemini 3.8 Flash (medium)	Google	40	-	1M	US$0.6	-	不可用		查看
Gemini 3.7 Flash (medium)	Google	40	-	1M	US$0.6	304	不可用		查看
Gemini 3.7 Flash (high)	Google	39	-	1M	US$0.6	308	不可用		查看
Gemini 3.7 Flash (low)	Google	37	-	1M	US$0.6	310	不可用		查看
Gemini 3.6 Flash (high)	Google	34	-	1M	US$0.6	209	不可用		查看
Gemini 3.5 Flash (medium)	Google	34	-	1M	US$1.3	270	不可用		查看
Gemini 3.8 Flash (low)	Google	33	-	1M	US$0.6	-	不可用		查看
Gemini 3.5 Flash (high)	Google	33	-	1M	US$1.3	204	不可用		查看
Gemini 3.1 Pro Preview	Google	30	-	1M	US$1.7	116	不可用		查看
Gemini 3 Pro Preview (high)	Google	28	-	1M	US$1.7	-	不可用		查看
Gemini 3 Flash Preview (Reasoning)	Google	26	-	1M	US$0.4	215	不可用		查看
Gemini 3.5 Flash (minimal)	Google	24	-	1M	US$1.3	212	不可用		查看
Gemini 3 Pro Preview (low)	Google	22	-	1M	US$1.7	-	不可用		查看
Gemini 3.5 Flash-Lite	Google	22	-	1M	US$0.3	337	不可用		查看
Gemma 4 31B (Reasoning)	Google	19	30.7B	256k	US$0.0	35	🤗	+8	查看
Gemini 3 Flash Preview (Non-reasoning)	Google	18	-	1M	US$0.4	220	不可用		查看
Gemma 4 26B A4B (Reasoning)	Google	17	25.2B 推理时启用 3.8B 个参数	256k	US$0.1	-	🤗	+6	查看
Gemini 2.5 Pro	Google	16	-	1M	US$1.3	139	不可用		查看
Gemini 3.1 Flash-Lite	Google	16	-	1M	US$0.2	320	不可用		查看
Gemini 2.5 Flash Preview (Sep '25) (Reasoning)	Google	15	-	1M	-	-	不可用	-	查看
Gemini 2.5 Pro Preview (Mar' 25)	Google	15	-	1M	-	-	不可用	-	查看
Gemini 2.5 Pro Preview (May' 25)	Google	15	-	1M	US$1.3	-	不可用		查看
Gemma 4 12B (Reasoning)	Google	14	12B	256k	US$0.1	114	🤗		查看
Gemma 4 31B (Non-reasoning)	Google	14	30.7B	256k	US$0.2	54	🤗	+6	查看
Gemma 4 26B A4B (Non-reasoning)	Google	13	25.2B 推理时启用 3.8B 个参数	256k	US$0.1	107	🤗	+6	查看
Gemini 2.5 Flash (Reasoning)	Google	13	-	1M	US$0.3	221	不可用		查看
Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)	Google	12	-	1M	-	-	不可用	-	查看
Gemini 2.5 Flash Preview (Reasoning)	Google	12	-	1M	-	-	不可用	-	查看
Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)	Google	10	-	1M	US$0.1	-	不可用		查看
Gemini 2.5 Flash (Non-reasoning)	Google	10	-	1M	US$0.3	226	不可用		查看
DiffusionGemma 26B A4B	Google	10	25.2B 推理时启用 3.8B 个参数	256k	-	-	🤗	-	查看
Gemini 2.0 Flash Thinking Experimental (Jan '25)	Google	9	-	1M	-	-	不可用	-	查看
Gemma 4 12B (Non-reasoning)	Google	9	12B	262k	US$0.1	124	🤗		查看
Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)	Google	9	-	1M	US$0.1	-	不可用		查看
Gemini 2.0 Flash (Feb '25)	Google	9	-	1M	-	-	不可用	-	查看
Gemma 4 E4B (Reasoning)	Google	9	8B 推理时启用 4.5B 个参数	128k	US$0.0	58	🤗		查看
Gemini 2.0 Pro Experimental (Feb '25)	Google	9	-	2M	-	-	不可用	-	查看
Gemini 2.5 Flash Preview (Non-reasoning)	Google	9	-	1M	-	-	不可用	-	查看
Gemini 2.5 Flash-Lite (Reasoning)	Google	9	-	1M	US$0.1	363	不可用		查看
Gemini 2.0 Flash (experimental)	Google	8	-	1M	US$0.0	-	不可用		查看
Gemini 1.5 Pro (Sep '24)	Google	8	-	2M	-	-	不可用	-	查看
Gemma 4 E2B (Reasoning)	Google	8	5.1B 推理时启用 2.3B 个参数	128k	-	-	🤗	-	查看
Gemma 4 E4B (Non-reasoning)	Google	7	8B 推理时启用 4.5B 个参数	128k	US$0.0	60	🤗		查看
Gemini 2.0 Flash-Lite (Feb '25)	Google	7	-	1M	-	-	不可用	-	查看
Gemini 2.0 Flash-Lite (Preview)	Google	7	-	1M	-	-	不可用	-	查看
Gemini 1.5 Flash (Sep '24)	Google	7	-	1M	-	-	不可用	-	查看
Gemini 2.5 Flash-Lite (Non-reasoning)	Google	7	-	1M	US$0.1	294	不可用		查看
Gemini 2.0 Flash Thinking Experimental (Dec '24)	Google	7	-	2M	-	-	不可用	-	查看
Gemma 4 E2B (Non-reasoning)	Google	6	5.1B 推理时启用 2.3B 个参数	128k	-	-	🤗	-	查看
Gemini 1.5 Pro (May '24)	Google	6	-	2M	-	-	不可用	-	查看
Gemini 1.5 Flash-8B	Google	6	8B	1M	-	-	不可用	-	查看
Gemini 1.5 Flash (May '24)	Google	6	-	1M	-	-	不可用	-	查看
Gemini 1.0 Ultra	Google	6	-	33k	-	-	不可用	-	查看
Gemma 3n E4B Instruct Preview (May '25)	Google	6	8.4B 推理时启用 4B 个参数	32k	-	-	🤗	-	查看
PALM-2	Google	5	-	8k	-	-	不可用	-	查看
Gemini 1.0 Pro	Google	5	-	33k	-	-	不可用	-	查看
Gemma 3 270M	Google	5	0.3B	32k	-	-	🤗	-	查看
Gemma 3 27B Instruct	Google	5	27.4B	128k	US$0.0	-	🤗	+3	查看
Gemma 3 4B Instruct	Google	5	4.3B	128k	US$0.0	-	🤗		查看
Gemma 3n E4B Instruct	Google	5	8.4B 推理时启用 4B 个参数	32k	-	-	🤗	-	查看
Gemma 3n E2B Instruct	Google	5	6.0B 推理时启用 2B 个参数	32k	US$0.0	-	🤗		查看
Gemma 3 1B Instruct	Google	5	1B	32k	US$0.0	-	🤗		查看
Gemma 3 12B Instruct	Google	4	12.2B	128k	US$0.0	-	🤗		查看
Gemini 3 Deep Think	Google	-	-	128k	-	-	不可用	-	查看

Google 模型：智能、性能与价格

智能

速度

成本

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Index vs. Cost per Intelligence Index Task

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

成本

Cost per Intelligence Index Task

Cost per Intelligence Index Task

速度与延迟

Output Speed

Output Speed

Model Performance Representation

能力得分

能力指数

Google 的所有发布

更多详情