跳到主要内容Skip to main content
← Automation

Relevance AIRelevance AI

Relevance AI(悉尼)Relevance AI (Sydney)
无代码No-code 多agentMulti-agent 云托管Cloud-hosted 销售运营自动化Sales operations automation partial lifecycle: active 核验Checked 2026-10-30

Relevance AI 主打「无代码搭 AI 员工团队」——多 agent 协作是它的叙事主轴,销售/运营自动化是主战场。 云托管、无自托管。

Relevance AI leads with “build an AI workforce with no code” — multi-agent collaboration is its narrative axis, sales/operations automation is the main battlefield. Cloud-hosted, no self-hosting.

适合与不适合When to use it, when not to

适合:无代码多 agent 工作流、销售/运营场景自动化。 不适合:要自托管;要代码级控制;任务简单到不需要多角色分工(先问这一句)。

A fit for: no-code multi-agent workflows, sales/operations scenario automation. Not a fit for self-hosting; code-level control; tasks simple enough not to need multi-role division of labor (ask that question first).

中文生态China availability 国产 / 国内可直接用 / 备案状态 · 快照domestic / direct access / CAC filing · snapshot 2026-10-21

出品属地Origin
国外出品foreign-made Relevance AI(澳大利亚,悉尼)。Relevance AI (Australia, Sydney).
国内使用Direct access
大陆可直连direct OK relevanceai.com / docs.relevanceai.com 直连实测 200(2026-10-21)。relevanceai.com / docs.relevanceai.com measured 200 direct (2026-10-21).
备案状态CAC filing
未见于备案名单not in filing list:未见于网信办公示名单(2026-10-21 检索);该产品未面向大陆公众提供服务,备案不适用。not in the CAC public registry (searched 2026-10-21); the product does not serve the mainland public, so filing does not apply.

备案信息以网信办《生成式人工智能服务已备案信息》公示名单为准(按批次滚动更新,本站只抄编号不解读);「大陆可直连」是本站从大陆网络直连实测的结果,不是厂商承诺。GitHub 类站点当日实测有波动,口径写在依据里。Filing status is quoted from the CAC public registry (updated batch by batch; this site copies the number and does not interpret it). "Direct access" is this site's own reachability measurement from a mainland network, not a vendor promise. GitHub-hosted sites fluctuated on the measurement day — the note says so.

固定坐标系Fixed coordinate system

8 个维度dimensions, 与同分区其他对象逐项可比each comparable item by item against others in the same section.

模型与开放条件Model & openness

平台接各家模型(OpenAI/Anthropic 等,官方文档线);具体清单未逐条核验。

The platform connects to each vendor's models (OpenAI/Anthropic etc., official docs line); the specific list has not been verified item by item.

运行位置Where it runs

纯云托管:agent 团队("AI workforce")跑在 Relevance 云端;无自托管形态。

Pure cloud hosting: the agent team ("AI workforce") runs on Relevance's cloud; no self-hosted form.

本地文件Local files

不承载本机文件运行时——数据、附件与工具产出都在平台内存储与流转; 要「AI 读写本机文件」的形态在这里不存在(如实记,不是官方不允许)。

It hosts no local-file runtime — data, attachments and tool outputs are stored and flow inside the platform; the "AI reads/writes local files" form does not exist here (recorded honestly, not "officially not allowed").

关机后的任务Tasks after shutdown

云托管常驻;官方主打「无代码搭 AI 员工团队」——多 agent 协作是叙事主轴。

Cloud-hosted and resident; the official pitch is "build an AI employee team with no code" — multi-agent collaboration is the narrative main axis.

工具与扩展Tools & extensions

工具/集成 + MCP(官方文档线含 MCP 条目,未逐条核验细节)。

Tools/integrations + MCP (the official docs line includes MCP entries; details not verified item by item).

上下文与记忆Context & memory

agent 间任务传递与共享指令是官方协作面;跨会话记忆未逐条核验。

Task passing between agents and shared instructions are the official collaboration surface; cross-session memory not verified item by item.

权限与限制Permissions & limits

团队/工作区权限面见官方文档线(角色与共享边界未逐条核验); 企业合规声明(SOC 2 类)未逐条核验——「未核验到」不等于「没有」。

Team/workspace permission surface is in the official docs line (roles and sharing boundaries not verified item by item); enterprise-compliance claims (SOC 2 type) not verified item by item — "not verified" does not mean "does not exist".

适合什么任务Best for

适合:无代码搭「多 agent 协作」工作流、销售/运营自动化场景。 不适合:要自托管;要代码级控制(归 framework)。

Fits: building "multi-agent collaboration" workflows with no code; sales/operations automation scenarios. Does not fit: self-hosting; code-level control (belongs to framework).

头号误解Top misconception

  • 「AI workforce」叙事下把多 agent 当万能——先核任务是否真需要多角色分工Treating multi-agent as a universal answer under the “AI workforce” narrative — first check whether the task truly needs multi-role division
  • 官网 terms 旧路径 404——条款走 docs 线读(2026-10-21 直抓核验)The old terms path on the site is 404 — read terms via the docs line (direct-fetched 2026-10-21)

未知项清单Known unknowns

  • 模型清单与切换粒度(未逐条核验)
  • MCP 支持细节(文档线有条目未逐条核验)
  • 官方 changelog 页现值(docs 子路径未逐一核验)
  • The model list and switching granularity (not verified item by item)
  • MCP support details (docs-line entries exist but not verified item by item)
  • The current official changelog page (docs sub-paths not verified one by one)

证据来源Evidence sources

判断可回到以下一手源复核。本站核验日Every judgement can be rechecked against the primary sources below. Verified on 2026-10-30,内容更新日content updated on 2026-10-21。

类型Type名称Name链接Link
docsRelevance AI · 官方文档(2026-10-21 直抓 200)Relevance AI · official docs (fetched directly 200 on 2026-10-21)https://docs.relevanceai.com
docsRelevance AI · 官方首页(2026-10-21 直抓 200)Relevance AI · official homepage (fetched directly 200 on 2026-10-21)https://relevanceai.com
changelogRelevance AI · 文档更新记录(changelog 入口)Relevance AI · documentation update records (changelog entry point)https://docs.relevanceai.com/changelog

实测记录Measured runs

本档未跑工作流搭建实测;结论只覆盖平台能力与权限边界。

This site has not completed measured runs. The closest test protocol is the code-side protocols/framework/_protocol.md (the tool-calling and credential-boundary sections are reusable; no workflow-platform-specific protocol exists on this site yet).

相关条目Related entries

  • 扣子 / CozeCoze同为「多 agent 协作」无代码平台:Relevance 偏销售运营英文市场,扣子偏国内分发链路。Both no-code "multi-agent collaboration" platforms: Relevance leans toward sales/operations in English markets, Coze leans toward domestic distribution chains.
  • CrewAICrewAI另一层 · 框架与运行时Other layer · Frameworks & runtimes

本页由This page is generated from the ai-handbook 数据层生成(CC BY 4.0)。data layer (CC BY 4.0). 方法论与坐标系定义见仓库内For the methodology and coordinate definitions, see METHODOLOGY.md.