OpenAI Web Search

OpenAI Web Search：Search Index 得分、性能与价格

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

比较 Search API

Search Index

OpenAI Web Search（medium）：Artificial Analysis Search Index

耗时

32.7s

OpenAI Web Search（medium）：每任务的平均模型耗时与搜索耗时

成本

$0.05

OpenAI Web Search（medium）：每任务的平均模型成本与 Search API 成本（美元）

与其他 Search API 相比，OpenAI Web Search 的 Search Index 得分处于平均水平，价格也较为合理。此外，其速度慢于平均水平。我们测试了一种变体：OpenAI Web Search（medium）。

OpenAI Web Search（medium）：Artificial Analysis Search Index 得分为 74，在 Search API 服务商中处于中位数水平（中位数：74）。

OpenAI Web Search（medium）：每项任务耗时 32.7s，慢于平均水平（中位数：31.6s）。

OpenAI Web Search（medium）：在 Search Index 上运行的平均每项任务成本为 $0.05（含候选模型；中等，中位数：$0.061）。

变体	OpenAI Web Search（medium）
候选模型	GPT-5.6 Luna (medium)

结果

在各项公开基准测试中启用搜索后的回答质量。

Artificial Analysis Search Index

DeepSearchQA F1、BrowseComp 准确率与 AA-Omniscience 准确率的等权平均值 · Higher is better

仅模型（33）

集成的第一方搜索工具与框架

The equal-weighted mean of each Search API provider's score on DeepSearchQA (average F1), BrowseComp (exact-answer accuracy), and AA-Omniscience (accuracy), shown on a 0-100 scale. Every result runs the same candidate answer model, harness, and settings. The only variable is the Search API provider.

The dashed line represents the candidate answer model, GPT-5.6 Luna (medium), run with no search tools and is considered the baseline. Dotted bars show the same model with an integrated first-party search tool and harness (OpenAI Web Search). See the Search API methodology for the full harness settings.

成本

每项基准测试任务和每次搜索查询的候选模型成本与 Search API 成本。

每任务成本

每任务的平均候选模型成本与 Search API 成本（美元） · Lower is better

仅模型（$0.0029）

集成的第一方搜索工具与框架

Total cost of one benchmark task, split into the candidate answer model's tokens (input, cached, reasoning, and output) and the Search API provider's list price for the searches the model chose to run. The model-only baseline runs no searches, so its cost is token cost alone. Model cost can differ per provider due to differences in returned payloads for a given query, increased numbers of searches performed (due to not getting the right results) as well as increased reasoning token usage when search results are of lower quality.

Artificial Analysis Search Index 与每任务成本

Artificial Analysis Search Index 与每项基准测试任务的模型与搜索总成本（美元）

Most attractive quadrant

Pareto line

延迟

每项基准测试任务和每次搜索查询的模型耗时与搜索耗时。

每任务耗时

每任务的平均推算模型耗时与实测搜索耗时 · 模型耗时为输出 token 数除以端到端输出速度，速度取候选模型自有接入点的数值，或各服务商的中位数 · Lower is better

仅模型（25.6s）

集成的第一方搜索工具与框架

The sum of model time and search time for one benchmark task. Model time is derived: answer and reasoning tokens per task divided by the model's end-to-end output speed, at its own endpoint or the median across providers. Search time is the measured time spent in web_search calls. A provider can be fast per call and still add more total time if the model searches against it more often.

For OpenAI Web Search, search time is an upper limit. It includes OpenAI processing time.

Artificial Analysis Search Index 与每任务耗时

Artificial Analysis Search Index 与每项基准测试任务的平均模型与搜索耗时（秒） · 模型耗时为输出 token 数除以端到端输出速度，速度取候选模型自有接入点的数值，或各服务商的中位数

Most attractive quadrant

Pareto line

For OpenAI Web Search, search time is an upper limit. It includes OpenAI processing time.

Token

每项基准测试任务的候选模型输入与输出 token 数，以及质量与 token 总用量的关系。

每任务 token 总数

每项基准测试任务中候选模型的平均输入、推理与回答 token 数 · Lower is better

仅模型（3k）

集成的第一方搜索工具与框架

Average total input, reasoning and answer tokens per task. Grader tokens are excluded.

Artificial Analysis Search Index 与每任务 token 总数

Artificial Analysis Search Index 与每项基准测试任务的候选模型 token 总数

Most attractive quadrant

Pareto line

Average total input, reasoning and answer tokens per task. Grader tokens are excluded.

搜索

候选模型在每项基准测试任务中执行的搜索查询次数。

每任务搜索查询次数

候选模型在每项基准测试任务中执行的平均搜索查询次数

The average number of web_search calls the candidate answer model makes for one benchmark task, averaged across the three benchmarks. The model decides when to search, so the count reflects how it works with each provider's results. A higher count is not necessarily worse, a provider may return results intentionally designed to encourage more, frequent searches. The model-only baseline runs no searches.

基准测试详情

OpenAI Web Search 的 Search Index、基准测试得分、每任务成本与耗时，以及仅模型基线

服务商									网站
OpenAI Web Search（medium）	74	41	78	74	72	$40.19	$9.40	32.7s	openai.com
仅模型	33	基线	45	17	38	—	$2.92	25.6s

Search $/1k and Model $/1k are the average cost of 1,000 benchmark tasks broken down into search and model token costs. The model-only baseline runs no searches.

For OpenAI Web Search, search time is an upper limit. It includes OpenAI processing time.

OpenAI Web Search：Search Index 得分、性能与价格

服务商摘要

Search Index

耗时

成本

比较摘要

技术规格

结果

Artificial Analysis Search Index

Artificial Analysis Search Index

Model-Only Baseline and First-Party Search

成本

每任务成本

Cost per Task

Model-Only Baseline and First-Party Search

Artificial Analysis Search Index 与每任务成本

Cost per Task

延迟

每任务耗时

Time per Task

Model-Only Baseline and First-Party Search

Artificial Analysis Search Index 与每任务耗时

Time per Task

Token

每任务 token 总数

Total Tokens per Task

Model-Only Baseline and First-Party Search

Artificial Analysis Search Index 与每任务 token 总数

Total Tokens per Task

搜索

每任务搜索查询次数

Search Queries per Task

基准测试详情

Cost per 1,000 Tasks

Time per Task