免费 / 开源主要作为本地模型的轻量级 CLI 运行时,而非功能齐全的聊天界面。
Open WebUI 记录了 804,534 月访问量,月度环比增长为 -2.8%,定价为免费增值。Roark 记录了 9,226 月访问量,增长率为 +23.1%,定价为付费。

Open WebUI is an open-source, self-hosted AI interface that lets individuals and organizations run any model — local or cloud — inside a workspace they fully control. It ships as a one-command install, supports Ollama, OpenAI-compatible and Anthropic endpoints, and extends through Python tools, functions and pipelines.

Roark is a QA and observability platform for voice AI agents. It simulates your agent against hundreds of realistic callers before launch, then scores every production call on 500+ audio-native metrics so you can catch failures, prove a fix and verify the improvement in live calls.
| 维度 / 指标 | ![]() | ![]() |
|---|---|---|
| 官方域名 | openwebui.com | roark.ai |
| 定价模式 | 免费增值 | 付费 |
| 月访问量 | 80.45万 (804,534) | 9226 (9,226) |
| 月度环比流量变化 | -2.3万 (-2.8%) | +1734 (+23.1%) |
| 主要市场与份额 | 🇺🇸 美国 (24.5%) | 🇺🇸 美国 (75.7%) |
| 直接访问 / 月 | 29.98万 | 4831 |
| 用户评分 | 4.4 | 4.4 |
| 支持的平台 | 网站 | 网站 |

Open WebUI is an open-source, self-hosted AI interface that lets individuals and organizations run any model — local or cloud — inside a workspace they fully control. It ships as a one-command install, supports Ollama, OpenAI-compatible and Anthropic endpoints, and extends through Python tools, functions and pipelines.
Connect local runtimes like Ollama alongside OpenAI-compatible and Anthropic APIs, and switch between them inside one interface.
RAG over your own documents, web search and citation-aware answers come standard, so responses are grounded in real sources.
Extend the platform with Python code to add custom tooling, preprocessing, and multi-step pipelines, then share them with others.
SSO, RBAC, audit logs, data residency, and on-premises, hybrid or air-gapped deployment for organisations with strict requirements.

Roark is a QA and observability platform for voice AI agents. It simulates your agent against hundreds of realistic callers before launch, then scores every production call on 500+ audio-native metrics so you can catch failures, prove a fix and verify the improvement in live calls.
Run your agent against hundreds of realistic personas, accents, background noise and edge cases before a customer ever dials in.
Red-team callers attempt prompt injection, jailbreaks and social engineering to check your agent holds policy under pressure.
Every live call is scored as it lands on more than 500 metrics, with issues filed, alerts fired and traces attached.
Purpose-built audio models measure pronunciation, emotion, vocal stress, pace, interruptions and dead air, not just transcript quality.
免费 / 开源主要作为本地模型的轻量级 CLI 运行时,而非功能齐全的聊天界面。

免费 / 开源Focuses on collaborative team AI workspaces and plugins, but shares similar open-source architectural goals.
专项更强强烈专注于注重隐私的文件聊天与本地 RAG 管道,但提供较少的通用聊天 UI 功能。
免费 / 开源提供可在离线时轻松运行本地模型的桌面原生应用程序,但缺乏 Open WebUI 以服务器为核心的多用户功能。

免费 / 开源面向生产级 AI 的智能体可观测性、评估与持续改进平台
免费 / 开源构建、模拟、评估并监控可自我改进的 AI 智能体