# Best open-source research & data agents

> The open-source research & data agents in the index — self-hostable, inspectable, and free to run. Ranked by real maintained adoption (GitHub stars and recent commit activity), never by sponsorship. Open source buys you control, privacy, and no vendor lock-in; the trade-off is that you host and maintain it yourself.

Right now it's led by firecrawl (160k stars), with Scrapling and TrendRadar close behind. Rankings shift as projects gain stars and ship commits.

23 open-source records, ranked by maintained GitHub adoption; sponsorship never affects the ranking. Browse on the web: https://agentsearchengine.app/best/open-source-research-agents

## Ranked

1. [firecrawl](https://agentsearchengine.app/agents/firecrawl) — The API to search, scrape, and interact with the web at scale. · Open source · Infrastructure · 160k★ · TypeScript · AGPL-3.0 · active 1 day ago · best for feeding agents clean web data: crawl, scrape, and convert to markdown via api
2. [Scrapling](https://agentsearchengine.app/agents/scrapling) — An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! · Open source · Infrastructure · 72k★ · Python · BSD-3-Clause · active 4 days ago · best for scraping sites that change layout or block bots, with selectors that self-repair
3. [TrendRadar](https://agentsearchengine.app/agents/trendradar) — AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts. · Open source · Agent · 61k★ · Python · GPL-3.0 · active 17 days ago · best for filtering trending news into push alerts on wechat, telegram, feishu, or slack
4. [BettaFish](https://agentsearchengine.app/agents/bettafish) — 微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。 · Open source · Agent · 42k★ · Python · GPL-2.0 · active 13 days ago · best for multi-agent sentiment analysis across social platforms and large comment datasets
5. [khoj](https://agentsearchengine.app/agents/khoj) — Your AI second brain. Self-hostable. Get answers from the web or your docs. Build custom agents, schedule automations, do deep research. · Open source · Agent · 36k★ · Python · AGPL-3.0 · active 1 day ago · best for a self-hostable ai second brain over your notes and documents
6. [storm](https://agentsearchengine.app/agents/storm) — An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations. · Open source · Agent · 31k★ · Python · MIT · active 10 months ago · best for generating wikipedia-style researched reports with citations from an open-source system
7. [FastGPT](https://agentsearchengine.app/agents/fastgpt) — FastGPT is a knowledge-based platform built on the LLMs, offers a comprehensive suite of out-of-the-box capabilities such as data processing, RAG retrieval, and visual AI workflow orchestration, · Open source · Platform · 29k★ · TypeScript · active today · best for building rag agents with visual workflow orchestration and built-in data processing
8. [gpt-researcher](https://agentsearchengine.app/agents/gpt-researcher) — An autonomous agent that conducts deep research on any data using any LLM providers · Open source · Agent · 29k★ · Python · Apache-2.0 · active 16 days ago · best for self-hosted autonomous research reports with your own api keys
9. [haystack](https://agentsearchengine.app/agents/haystack) — Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. · Open source · Framework · 26k★ · Python · Apache-2.0 · active 2 days ago · best for modular python pipelines for production rag, search, and agent workflows
10. [DeepResearch](https://agentsearchengine.app/agents/deepresearch) — Tongyi Deep Research, the Leading Open-source Deep Research Agent · Open source · Agent · 20k★ · Python · Apache-2.0 · active 5 months ago · best for running alibaba's tongyi deepresearch model locally for deep research tasks
11. [WeKnora](https://agentsearchengine.app/agents/weknora) — Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki. · Open source · Platform · 19k★ · Go · active 2 days ago · best for enterprise document understanding with rag, a react agent, and auto-generated wikis
12. [local-deep-research](https://agentsearchengine.app/agents/local-deep-research) — ~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google,...). 10+ search engines - arXiv, PubMed, your private documents. · Open source · Agent · 8.8k★ · Python · MIT · active today
13. [MiroThinker](https://agentsearchengine.app/agents/mirothinker) — MiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively. · Open source · Agent · 8.4k★ · Python · Apache-2.0 · active 28 days ago
14. [deep-searcher](https://agentsearchengine.app/agents/deep-searcher) — Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python. · Open source · Agent · 8.0k★ · Python · Apache-2.0 · active 8 months ago
15. [unstract](https://agentsearchengine.app/agents/unstract) — LLM-Driven Extraction of Unstructured Data — Built for API Deployments & ETL Pipeline Workflows · Open source · Platform · 7.1k★ · Python · AGPL-3.0 · active 1 day ago
16. [DeepAnalyze](https://agentsearchengine.app/agents/deepanalyze) — DeepAnalyze is the first agentic LLM for autonomous data science. 你的AI数据分析师，自动分析大量数据，一键生成专业分析报告！ · Open source · Agent · 4.4k★ · Python · MIT · active 1 month ago
17. [ms-agent](https://agentsearchengine.app/agents/ms-agent) — MS-Agent: a lightweight framework to empower agentic execution of complex tasks · Open source · Framework · 4.4k★ · Python · Apache-2.0 · active 3 days ago
18. [Sidekick](https://agentsearchengine.app/agents/sidekick) — A native macOS app that allows users to chat with a local LLM that can respond with information from files, folders and websites on your Mac without installing any other software. · Open source · Agent · 3.3k★ · Swift · MIT · active 2 months ago
19. [MiroFlow](https://agentsearchengine.app/agents/miroflow) — Top-1 on 5+ benchmarks | Web UI | Supports MiroThinker, Claude, Kimi, OpenAI · Open source · Framework · 3.1k★ · Python · Apache-2.0 · active 28 days ago
20. [Adala](https://agentsearchengine.app/agents/adala) — Adala: Autonomous DAta (Labeling) Agent framework · Open source · Framework · 1.6k★ · Python · Apache-2.0 · active 6 days ago
21. [ToolUniverse](https://agentsearchengine.app/agents/tooluniverse) — Democratizing AI scientists with ToolUniverse · Open source · SDK / library · 1.6k★ · Python · Apache-2.0 · active today
22. [Auto-Deep-Research](https://agentsearchengine.app/agents/auto-deep-research) — "Your Fully-Automated Personal AI Assistant" · Open source · Agent · 1.6k★ · Python · active 9 months ago
23. [TrustRAG](https://agentsearchengine.app/agents/trustrag) — TrustRAG：The RAG Framework within Reliable input,Trusted output · Open source · Framework · 1.3k★ · Python · active 6 months ago

## FAQ

**What are the best open-source research & data agents?**
The projects listed below, ranked by maintained adoption — GitHub stars plus recent activity — so you see what's genuinely used and still maintained, not abandoned.

**Are open-source research & data agents free?**
The software is free to run and self-host. You still pay for the infrastructure you run it on and any model or API usage it makes.

**Open-source or commercial research & data agents — which should I choose?**
Pick open source for control, privacy, and no lock-in when you're happy to host and maintain it; pick a commercial product for a managed, supported experience. You can compare both in the full research & data agents category.

---
Source: [Agent Search Engine](https://agentsearchengine.app) — https://agentsearchengine.app/best/open-source-research-agents