OpenAgents
Open platform for language agents with data analysis, web browsing, and plugin capabilities.
Overview
OpenAgents is an open-source platform providing three specialised AI agents: a Data Agent (for data analysis using Python and SQL), a Plugins Agent (using web browsing, weather, search and other tools), and a Web Agent (for web browsing and task execution). Released by researchers at The University of Hong Kong and Sea AI Lab, it aims to make language agent development and deployment more accessible.
The Data Agent can analyse datasets uploaded by users, write and execute Python code for statistical analysis, create visualisations, and explain findings in plain English — comparable to OpenAI's Code Interpreter but open source. The Plugins Agent accesses real-time information through a set of customisable tools including weather, stock prices, and search.
OpenAgents is designed to bridge the gap between research agent frameworks (which require technical setup) and commercial agents (which lack customisability). The platform runs entirely in the browser with minimal setup and is aimed at data scientists, researchers, and developers who want to experiment with capable agent configurations without writing framework code.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 8 May 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
Use cases
ASEAN Perspective
OpenAgents in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
OpenAgents from XLang Lab is a research-oriented open-source platform demonstrating three agent types — a data agent, a plugins agent and a web-browsing agent — as an open counterpart to ChatGPT's plugin/agent experience. It is valuable as a reference implementation and study object for how to build host-and-use agent systems.
It suits researchers and developers exploring agent architectures who want a working, inspectable codebase. Caveats: it is an academic project, not a maintained product — expect rougher edges, slower updates and limited support compared with commercial agent stacks, and it has been overtaken by faster-moving frameworks. Free and open-source, English/global, but better as a learning resource than a production foundation.
Notable facts
- OpenAgents was built collaboratively between researchers at The University of Hong Kong and Sea AI Lab, a Singapore-headquartered research organisation — making it one of the most prominent AI agent projects to emerge from Southeast Asia.
- The Data Agent can analyse a CSV file, identify statistical anomalies, write visualisation code, and explain the findings in plain English — all in a single conversation.
- OpenAgents was released with a fully functional demo at openagents.xlang.ai requiring no code or account to try.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including OpenAgents's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with OpenAgents unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to OpenAgents directly →
Spotted something out of date? Suggest an update →
Alternatives to OpenAgents
More in Agents & Automation
