← Automation

FlowiseFlowise

FlowiseAI, Inc.

开源Open source Apache2.0混合Mixed Apache-2.0 画布Canvas LLM应用LLM applications RAGRAG 自托管Self-hosted partial lifecycle: active 核验Checked 2026-10-30

Flowise 是「画布搭 LLM 应用」的开山一代（LangChain 组件族的可视化封装）：RAG/agent 流原型极快。两个诚实标注：许可是 Apache 2.0 + enterprise 商业区混合；2026-08 后仓库推送放缓——热度与维护节奏要看快照。

Flowise is the founding generation of “build LLM apps on a canvas” (a visual wrapper over LangChain components): RAG/agent-flow prototyping is extremely fast. Two honest notes: the license is Apache 2.0 + an enterprise commercial zone; repo pushes slowed after 2026-08 — momentum is a snapshot fact.

适合与不适合When to use it, when not to

适合：LLM 应用（RAG、agent 流）快速原型、要自托管、要读源码。不适合：要企业功能开箱（SSO 等在商业区）；要长期活跃发版承诺（按快照口径）。

A fit for: fast LLM-app prototyping (RAG, agent flows), self-hosting, reading the source. Not a fit for out-of-the-box enterprise features (SSO in the commercial zone); expecting a long active-release cadence (snapshot caveat).

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　FlowiseAI, Inc.（美国）。FlowiseAI, Inc. (US).
国内使用Direct access: 大陆可直连direct OK　docs.flowiseai.com 直连实测 200（2026-10-21）。docs.flowiseai.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: 画布节点接各家 LLM/嵌入/向量库（LangChain 组件族为底）；provider 覆盖面以官方 docs 为准，未逐条核验。
Canvas nodes connect to each vendor's LLMs/embeddings/vector stores (built on the LangChain component family); provider coverage follows the official docs, not verified item by item.
运行位置Where it runs: 开源自托管（npm/Docker）+ 官方云两条线；GitHub 55,513★（API 核验 2026-10-21；最近推送 2026-08-13—— 节奏放缓是事实，如实记）。
Open-source self-hosting (npm/Docker) + official cloud, two lines; GitHub 55,513★ (API-verified 2026-10-21; last push 2026-08-13 — the slowing cadence is a fact, recorded honestly).
本地文件Local files: 自托管实例数据落自有环境；chatflow 上传文件处理在实例内。
Self-hosted instance data lands in your own environment; chatflow uploaded-file handling happens inside the instance.
关机后的任务Tasks after shutdown: 自托管实例常开即承载 API 服务与队列；官方给 queue 模式（细节未逐条核验）。
A continuously running self-hosted instance carries API serving and queues; the official docs provide a queue mode (details not verified item by item).
工具与扩展Tools & extensions: 节点生态含工具/检索/agent 流；MCP 支持状态未逐条核验。
The node ecosystem includes tools/retrieval/agent flows; MCP support status not verified item by item.
上下文与记忆Context & memory: chatflow 的记忆节点（buffer/窗口等 LangChain 记忆组件）是官方上下文面；跨应用记忆未逐条核验。
Chatflow memory nodes (buffer/window and other LangChain memory components) are the official context surface; cross-app memory not verified item by item.
权限与限制Permissions & limits: 社区版基础鉴权；SSO/工作区隔离等在 enterprise 目录（商业许可区）。
Community edition has basic auth; SSO/workspace isolation etc. live in the enterprise directory (commercial-license zone).
适合什么任务Best for: 适合：画布搭 LLM 应用（RAG/agent 流）起步快、要自托管。不适合：要企业功能开箱；要活跃发版节奏（2026-08 后推送放缓——建档快照口径）。
Fits: fast starts building LLM apps on a canvas (RAG/agent flows), self-hosting. Does not fit: out-of-the-box enterprise features; an active release cadence (pushes slowed after 2026-08 — dossier-snapshot framing).

头号误解Top misconception

把「Apache 2.0 主体」读成全仓——enterprise 目录是商业区Reading “Apache 2.0 at the core” as whole-repo — the enterprise directory is the commercial zone
与 2024 年的星数/热度印象对齐——发版节奏已放缓，看仓库最近推送Aligning with 2024 star-count/hype impressions — release cadence has slowed; check the repo's last push
把画布原型当生产级运行时——原型快是它的强项，生产化要自己补Treating canvas prototypes as production runtimes — prototyping is its strength; production hardening is on you

未知项清单Known unknowns

provider/节点全清单（未逐条核验）
MCP 支持状态（未核验到官方文档）
发版节奏放缓的官方说明（未见——只记仓库事实，不猜原因）

The full provider/node catalog (not verified item by item)
MCP support status (no official documentation found)
An official explanation for the slowing release cadence (none found — only repository facts recorded, no guessing at causes)

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
docs	FlowiseAI/Flowise · LICENSE.md（Apache 2.0 + enterprise 商业区，2026-10-21 直抓核验）FlowiseAI/Flowise · LICENSE.md (Apache 2.0 + enterprise commercial zone, direct-fetched & verified 2026-10-21)	https://github.com/FlowiseAI/Flowise/blob/main/LICENSE.md
repo	FlowiseAI/Flowise · 仓库（55,513★，最近推送 2026-08-13，GitHub API 核验 2026-10-21）FlowiseAI/Flowise · repository (55,513★, last push 2026-08-13, verified via GitHub API 2026-10-21)	https://github.com/FlowiseAI/Flowise
docs	Flowise · 官方文档Flowise · official docs	https://docs.flowiseai.com
changelog	FlowiseAI/Flowise · Releases（changelog）FlowiseAI/Flowise · Releases (changelog)	https://github.com/FlowiseAI/Flowise/releases

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).