A question goes in; a cited, long-form report comes out - GPT Researcher is an open-source autonomous research agent. A planner agent decomposes the query into sub-questions, execution agents crawl 20+ web sources in parallel with JavaScript-enabled scraping, and a publisher aggregates findings into a 2,000+ word report with inline citations, exportable to PDF, Word, and Markdown. The Deep Research mode extends this recursively: each result yields follow-up questions that are explored to configurable breadth and depth in a tree pattern, while accumulated learnings, citations, and visited URLs are shared across branches. It also researches local documents (PDF, CSV, Word) alongside the web. LLM and search providers are pluggable, including OpenAI, Anthropic, Google, DeepSeek, and Ollama for models, and Tavily, Google, Bing, DuckDuckGo, and SearXNG for retrieval. It ships as a Python package, a FastAPI server with web frontend, a Docker image, and an MCP server for use inside Claude or Cursor. MIT-licensed.

Execute comprehensive multi-angle investigations across scientific journals and web repositories using Local Deep Research, an autonomous intelligence engine designed for researchers, analysts, and knowledge workers. Users can submit complex analytical questions through a self-hosted web console to initiate agentic research cycles powered by LangGraph workflows. The application dynamically queries search providers including SearXNG, arXiv, PubMed, Wikipedia, and Google, iteratively evaluating discovered content to formulate targeted follow-up queries. Discovered academic papers and web sources can be indexed into an encrypted personal library to combine internal documentation with external search intelligence. Reports are synthesized with structured outlines, section headings, and verifiable academic citations that link directly to source references. Individual research sessions, system logs, and user credentials remain isolated in per-user SQLCipher databases protected by AES-256 encryption. Automated subscriptions deliver scheduled news digests and topic monitoring summaries directly to connected notification channels. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Live web search plus an LLM of your choice: Farfalle is an open-source, self-hosted answer engine in the Perplexity mold. Queries route through one of several search providers - self-hosted SearXNG for a fully independent stack, or Tavily, Serper, and Bing APIs - and the model composes a cited answer from the retrieved results. Model flexibility is the core design: run llama3, mistral, gemma, or phi3 locally through Ollama for zero per-query cost and full privacy, use cloud models like GPT-4o or Groq-hosted Llama 3 for speed, or route to any provider via LiteLLM. An Expert Search mode uses an agent that plans a multi-step search strategy and executes it for harder questions, and chat history keeps prior research sessions available. The stack is a Next.js and shadcn/ui frontend over a FastAPI backend with Redis rate limiting, shipped as a pre-built Docker image. A browser search-engine entry pointing at your instance makes it the default search from the address bar. Paired with SearXNG and Ollama, the whole pipeline runs with no external API at all.

Perplexity's answer-engine experience, self-hostable and open-source: Morphic searches the web and writes cited answers. Instead of returning a list of links, it searches the web, reads the sources, and generates a complete answer with inline numbered citations. The generative UI streams rich components, source cards with thumbnails, image grids, syntax-highlighted code, and LaTeX math, rather than plain markdown. Quick mode answers fast; Adaptive mode runs deeper multi-step research. Search backends are pluggable: the Docker Compose bundle ships with a private SearXNG instance so no search API key is required, and Tavily, Brave, and Exa are supported alternatives. LLM providers include OpenAI, Anthropic, Google, Ollama, and any OpenAI-compatible endpoint, with per-mode model mapping - fast, cheap models for quick searches, stronger models for adaptive research, tuning the cost-quality trade-off per query type. An inspector panel exposes tool execution during multi-step research, and AI-suggested follow-up questions keep an investigation moving. Chat history persists in PostgreSQL, results are shareable by URL, file uploads feed context into queries, and optional Supabase authentication adds multi-user or guest access. Because the default search path is your private SearXNG instance, research topics never hit a commercial search API - and with local Ollama models the marginal cost of a query approaches zero. Built with Next.js, TypeScript, and the Vercel AI SDK under Apache 2.0.

DeerFlow 2.0 is ByteDance's ground-up rewrite that transforms what began as a Deep Research framework into a batteries-included super agent harness handling tasks lasting minutes to hours autonomously. Built on LangGraph and LangChain, the runtime orchestrates a lead agent that plans, decomposes, and delegates to sub-agents executing in isolated Docker or Kubernetes sandboxes with persistent filesystem access, while an extensible skills system lets developers add capabilities as Python functions or MCP servers with OAuth token flows. The harness ships with long-term memory using persistent event loops with per-agent isolation, scheduled task execution via cron, context engineering with manual compaction, and a web UI at localhost:2026 for interactive sessions. Model support spans OpenAI GPT-4o/GPT-5, Anthropic Claude via OAuth, Google Gemini, DeepSeek, Qwen via vLLM, and OpenRouter-compatible gateways with per-model pricing configuration. IM channel integration connects Telegram, Slack, Feishu/Lark, Discord, WeChat, WeCom, DingTalk, and Buzz without requiring a public IP — all using long-polling or WebSocket transports. Production deployment uses Docker Compose with nginx reverse proxy, PostgreSQL or SQLite persistence, Redis stream bridges for multi-worker scaling, and lease-based run ownership with automatic orphan recovery. The terminal workbench TUI enables headless operation. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Accepted as a NeurIPS 2025 Spotlight paper and rapidly approaching 6,000 GitHub stars, AI-Researcher from the Hong Kong University Data Science Lab delivers the first fully autonomous scientific research system — a multi-agent platform that takes a list of reference papers and returns a complete research contribution with working code, validated experiments, and a formatted academic manuscript. The pipeline orchestrates five distinct phases: a Resource Collector systematically gathers materials from arXiv, IEEE Xplore, ACM Digital Library, Google Scholar, GitHub, and Hugging Face; an Idea Generator performs gap analysis against semantic embeddings to produce 3-5 novel hypotheses with feasibility scores; an Algorithm Designer transforms concepts into functional implementations; a Validation Engine automates testing, performance evaluation, and iterative optimization; and a Manuscript Creator generates polished full-length papers with figures, tables, and citations. The Gradio-based web GUI provides intuitive tabs for environment configuration, example selection, and real-time monitoring of research progress, while the production deployment at novix.science offers immediate browser access without local setup. Scientist-Bench provides a standardized benchmark comprising state-of-the-art papers across diverse AI research domains for evaluating autonomous research capabilities. The system supports multiple LLM providers including OpenAI, Anthropic, Google Gemini, and OpenRouter models with per-task routing for cost optimization. Deploy via Python with pip dependencies or Docker containerization. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Positioned as the open-source NotebookLM alternative for AI agents, SurfSense delivers a live web research platform where your agents access structured data from Reddit, YouTube, Instagram, TikTok, Amazon, Walmart, Google Maps, Google Search, Indeed, and any page on the open web through one REST API or MCP server. Scheduled and event-triggered agents transform findings into briefs, alerts, podcasts, and presentations, while a built-in knowledge base keeps every discovery searchable with Perplexity-style cited answers using hybrid semantic and full-text search powered by PostgreSQL with pgvector. Upload PDFs, Office documents, images, and audio files, or sync Google Drive, OneDrive, and Dropbox — 50+ file formats supported with AI file sorting that auto-organizes documents by source, date, and topic. The MCP server exposes scrapers, knowledge base, and workspaces as native tools for Claude, Cursor, and any MCP-compatible agent. Cross-country proxy rotation handles Reddit, TikTok, and Google Search scraping with geo-aware sticky sessions and captcha-aware anti-bot handling. The platform features collaborative chats, multi-format document export, git-native knowledge base with Open Knowledge Format export, and a desktop quick-ask panel with global shortcut. Docker Compose deployment manages nine services including Caddy proxy, PostgreSQL, Redis, FastAPI backend, Celery workers, zero-cache real-time sync, and Next.js frontend with automatic Watchtower updates. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Perplexity's search experience without Perplexity: Vane deploys Perplexica, an open-source AI answer engine built as the self-hosted alternative. Instead of returning a page of links, it reads your question, searches the live web through the SearxNG metasearch engine, and composes a direct answer with cited sources. Retrieval quality comes from embeddings and similarity search: fetched pages are re-ranked against the query so the model answers from the most relevant passages rather than whatever ranked first. Two query modes cover different needs - Normal mode runs a straightforward web search, while Copilot mode generates multiple reformulated queries and actively pulls content from top matches for harder questions. Focus modes specialize retrieval for academic papers, YouTube, Reddit discussions, Wolfram Alpha calculations, or the general web. The answering model is your choice: OpenAI-compatible APIs or fully local LLMs such as Llama 3 and Mixtral through Ollama, which keeps queries entirely on your infrastructure. Because SearxNG pulls live results, answers reflect current information, and no search history is tracked.

Scira is an AI search engine that breaks complex research questions into sub-tasks, retrieves grounded answers from live sources, and cites every claim with clickable links to the original evidence. Formerly known as MiniPerplx and now backed by nearly 12,000 GitHub stars, the platform implements agentic research planning where language models decompose queries, select appropriate tools, retrieve information in parallel, and synthesize responses with inline citations. Seventeen distinct search modes cover Web, Academic, Extreme, X (Twitter), Reddit, GitHub, Stocks, Crypto, YouTube, and Prediction markets, each activating a curated subset of the 28 built-in tools. The Extreme mode deploys a deep research agent with multi-step planning, parallel source retrieval, code execution in sandboxed Daytona environments, and R2 artifact storage for generated charts. Model routing spans 50+ AI providers through a unified configuration layer supporting OpenAI, Anthropic, Google, xAI Grok, and custom models, with per-mode model assignment based on task requirements. The web search tool supports Exa, Tavily, and Firecrawl as interchangeable providers via a Strategy pattern. Lookouts schedule recurring research agents on cron schedules that monitor topics, track changes, and email summaries using Upstash QStash for reliable execution. PostgreSQL stores chat history, user preferences, and custom instructions while Redis handles caching and rate limiting. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

OpenNotebookLM is a self-hosted research workspace that reads your sources with you, answering questions only from the material you gave it and showing the passage behind every claim. Import papers, spreadsheets, recordings, web pages, PDF links, YouTube videos or playlists, Markdown folders, Obsidian vaults, and Git repositories, then select which sources the chat may use as evidence. Stream answers with follow-up questions, hover any numbered citation to inspect the supporting passage and page, and let the app flag citations that look unsupported before you trust them. The Studio turns the same sources into cited reports in briefing, study guide, FAQ, or timeline formats, expandable mind maps you can pan and question, two-host audio overviews in deep dive, brief, critique, or debate styles, narrated slides exportable to PowerPoint, and tables whose cells carry their own citations. Generate flashcards and timed quizzes from what you read, mark cards for review, and resume saved progress later. Save cited answers and Studio outputs as notes, convert notes back into searchable sources, and export almost anything: conversations, projects, mind maps as SVG, flashcards as Anki text, audio as MP3. Share projects with viewer or editor roles, keep conversations private per person, and bring Claude, OpenAI, any OpenAI-compatible provider, or a local Ollama model, with extraction, OCR, Whisper transcription, and Kokoro audio synthesis running on your own server. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

The most feature-complete open-source alternative to Google's NotebookLM — a self-hosted research platform where you upload PDFs, videos, audio files, and web pages into organized notebooks, then chat with your content, generate multi-speaker podcasts, and run semantic search across everything without sending a single byte to Google's servers. The podcast engine supports 1-4 fully customizable speakers with backstories, personalities, and expertise profiles, generating professional audio dialogue through OpenAI, ElevenLabs, Google TTS, or completely local text-to-speech via Kokoro for maximum privacy. Content processing uses token-based chunking with RAG-powered retrieval grounded in your uploaded sources, while both full-text keyword search and semantic vector search via SurrealDB enable conceptual discovery across all notebooks. The 18+ supported AI providers include OpenAI, Anthropic, Google Gemini, Groq, Ollama, LM Studio, and more — configurable per task so you can route cheap models to summarization and powerful models to analysis. Content transformations extract insights, generate summaries, create study guides, and produce structured outputs from any source material. The MCP integration connects Open Notebook to Claude Desktop, VS Code, and other MCP clients for seamless workflow integration. A full REST API on port 5055 enables complete automation of notebook management, source upload, and podcast generation. Deploy via Docker Compose with the application container, SurrealDB v2 on RocksDB, and optional TTS containers. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Insights LM gives you a NotebookLM-style research assistant that runs on your own infrastructure and answers strictly from the documents you feed it. Build a notebook for a project, drop in PDFs, pasted text, website links, and audio files, then ask questions in plain language and get answers grounded in those sources rather than the open internet. Every answer carries citations you can click to jump straight to the passage it came from, so verifying a claim takes seconds instead of a fresh search. Findings worth keeping become notes saved alongside the notebook, building a private knowledge base that grows with your research. When reading is too slow, generate an audio overview and listen to two hosts discuss your material like a podcast episode, perfect for absorbing a long report on a walk. Teams use it to interrogate internal documentation, onboard new hires against company knowledge, and brief themselves before meetings without uploading confidential files to a third-party service. Because the whole stack is yours, you choose which AI models do the reading, including local models for fully private work. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

SourceWeft is an AI workstation you host yourself, where agents, your documents, and your model keys work in one place instead of scattering across chat tabs you do not control. Describe a job and a lead agent breaks it down, handing research, exploration, and execution to focused sub-agents that share the same working context and files. Agents draw on a skill library covering presentations, HTML pages, images, meeting summaries, and study guides, and you can add your own skills so good workflows become repeatable. MCP servers, web search, and web extraction connect agents to outside tools, while approved commands run inside a sandboxed environment that cannot touch the rest of your server. Feed it PDFs, web pages, notes, YouTube videos, Notion, Drive, Gmail, or Slack, then follow citations from any answer back to the source material it came from. Bring your own keys for the chat and indexing models you prefer, organize the team into workspaces with roles and access controls, and pick up any conversation from the web app or desktop clients. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Chat with your own documents: AnythingLLM, from Mintplex Labs, wraps retrieval-augmented generation (RAG) in an open-source application anyone can run. You organize content into workspaces, each an isolated namespace with its own documents, vector embeddings, chat history, and settings, so one instance can hold several separate knowledge bases. Upload PDFs, DOCX, TXT, and other formats, or scrape web pages; the built-in collector parses and chunks them into a vector database (LanceDB by default, with Pinecone, Chroma, Qdrant, and others supported). Answers cite their source documents. It works with both cloud LLMs (OpenAI, Anthropic, Gemini) and local ones via Ollama or LM Studio, and the embedding model is separately configurable. Beyond RAG chat, it includes AI agents that can browse the web and run tools, an embeddable chat widget for your website, a developer API, and multi-user mode with admin, manager, and default roles plus per-workspace access control. Context assembly is smarter than naive RAG: pinned documents, attached files, vector search hits, and recent chat history are combined under a token budget so the model's context window is filled efficiently, and each workspace supports multiple independent conversation threads against the same knowledge base. Because the embedding model, vector store, and chat LLM are all independently swappable, you can move between providers without re-ingesting a single document. The stack is Node.js with a React frontend, MIT-licensed.

Large language models get a polished front end that can run fully offline: Open WebUI is the self-hosted front end of choice. It talks to local model runners, primarily Ollama, and to any OpenAI-compatible API, so LM Studio, vLLM, Groq, Mistral, OpenRouter, and cloud providers all plug into the same chat interface and can be mixed per conversation. RAG is built in: upload files to knowledge bases or reference them in chat with the # command, backed by a choice of nine vector databases (ChromaDB and PGVector officially maintained) and multiple extraction engines including Tika and Docling, with hybrid BM25-plus-vector search and cross-encoder reranking. Web search results from providers like SearXNG, Brave, and Tavily inject directly into conversations. Extensibility comes from Python tools and functions that run inside the chat, a Pipelines plugin framework, and native MCP support. Multi-user features include RBAC, SSO, and group permissions, and the instance itself exposes an OpenAI-compatible API your own apps can call.

With over 164,000 GitHub stars and one of the fastest-growing open-source projects in the AI tooling ecosystem, Firecrawl is the web context API that turns any website into clean markdown, structured JSON, or screenshots optimized for large language models. The Scrape endpoint converts a single URL into LLM-ready output with approximately 67% fewer tokens than raw HTML, handling JavaScript rendering, rotating proxies, anti-bot bypasses, and dynamic content extraction with zero configuration. The Crawl endpoint recursively scrapes entire websites from a single request with configurable depth, URL filters, and concurrent page limits. The Map endpoint discovers all URLs on a domain instantly for sitemap generation. The Search endpoint performs web searches and returns full page content from results. The Interact endpoint scrapes a page then continues working with it — clicking buttons, filling forms, and extracting dynamic content using AI prompts or code. The Agent endpoint provides autonomous web data gathering where users describe what they need in plain English. SDKs are available for Python, Node.js, Go, Rust, Ruby, PHP, Java, C#/.NET, and Elixir, with an MCP server for connecting to any AI agent or MCP client. Self-hosting deploys via Docker Compose and requires Redis and a Playwright-based browser service for JavaScript rendering. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Up to 280 search services - Google, Bing, DuckDuckGo, Brave, Qwant, Startpage - aggregated without tracking or profiling: SearXNG is a privacy-respecting metasearch engine (AGPL-3.0, successor to Searx). Your instance queries the upstream engines on your behalf: your IP address, cookies, and search history never reach them, tracker parameters are stripped from result URLs, and an optional image proxy fetches thumbnails server-side so result pages leak nothing. It can even route outbound queries through Tor for full anonymity. Search is organized into categories - general, images, videos, news, maps, music, IT, science, files - with bang shortcuts for targeting specific engines, and every source can be enabled, disabled, or weighted per category in settings.yml. A plugin system adds calculators, hash tools, tracker removal, and unit conversions inline, and preferences (themes, safe search, languages, engine selection) persist in cookies rather than server-side accounts. The real argument for running your own instance rather than trusting a public one is control: you decide the logging policy (none), the engine mix, rate limiting, and who gets access - making it the default search backend for browsers, families, and teams that want Google-quality results without the profile.