Search AI apps

Alibaba's AI super-app for search, writing, research and document tasks.

Quark is Alibaba's flagship consumer AI app, positioned as an 'AI super-box' that folds chat, deep search, deep research, AI writing, slide and image generation, document tools and cloud storage into a single interface. It identifies the intent behind a request, then plans and executes multi-step tasks. Quark runs on Alibaba's Qwen/Tongyi models and is offered across web, mobile and desktop.

Open-source RAG engine with deep document understanding.

RAGFlow is an open-source retrieval-augmented generation engine that turns complex documents—PDFs, slides, spreadsheets, scans, and web pages—into grounded, citation-backed context for LLMs. Its DeepDoc parser and hybrid vector plus full-text search aim for reliable, hallucination-resistant question answering, and it now bundles an agent-orchestration layer. Self-host the Apache-2.0 engine or use the managed cloud.

Rerankers and embeddings that sharpen AI retrieval.

ZeroEntropy builds specialized models — rerankers, embeddings, and custom retrieval models — that improve search accuracy in RAG pipelines and agentic apps. Its zerank rerankers reorder retrieved results so the most relevant passages reach the model, reportedly beating Cohere and Gemini rerankers at lower cost. The models are served through a single, latency-optimized API with Python and TypeScript SDKs, and the smaller zerank model is released as open weights.

Retrieval-specialized embedding and reranking models for RAG.

Voyage AI builds retrieval-specialized embedding and reranking models served via API to ground LLM applications. Its voyage-3 series of text and code embeddings, domain-tuned variants, and rerank models (e.g. rerank-2.5) are aimed at higher RAG accuracy than general-purpose embeddings. Now part of MongoDB, its models are being woven into Atlas Vector Search while the standalone API continues to operate.

Managed RAG-as-a-service — the context engine for AI agents and apps.

Ragie is a fully managed retrieval-augmented-generation platform. It ingests data through native connectors like Google Drive and Notion, parses multimodal content (PDFs, images, audio, video), and serves hybrid vector + keyword + summary retrieval over an API and MCP server. Developers add accurate, grounded context to LLM apps without building their own ingestion and retrieval pipeline.

Ad-free, privacy-first AI search that answers instead of returning links.

Andi is a generative AI search engine that gives direct, summarized answers in a conversational interface rather than a page of links. It runs its own search index and a proprietary content-understanding backend, layering language models on top to assess source meaning and quality. It positions itself as private and ad-free, with the core search staying free.

AI answer engine for developers — technical search with code and citations.

Phind is a conversational search engine built for developers: it combines live web results with generative AI to answer technical questions, returning explanations, code snippets, and cited sources. It uses multi-step reasoning over documentation and the web, and integrates into the editor via a VS Code extension. A free tier covers everyday use, with Plus and Pro plans adding stronger models, longer context, and higher limits.

Find any scene in your video and multimodal library.

Mixpeek is a multimodal retrieval API for searching across video, images, audio, and documents with natural language. It extracts and indexes structured features — faces, scenes, transcripts, OCR, and embeddings — over object storage like S3, GCS, and R2, then runs hybrid dense, sparse, and BM25 search with reranking. Cross-modal joins let a single query combine signals such as faces, spoken phrases, and on-screen text.

Real-time search and deep research API for AI agents.

Valyu is a search and retrieval API built for AI agents and LLMs. Its DeepSearch returns full extracted, structured content — not just links — from the open web plus proprietary and paywalled sources like academic papers, regulatory filings, patents, and financial data. A separate DeepResearch API runs asynchronous, multi-step research across those sources and synthesizes the results into reports and structured data.

Enterprise RAG platform for building grounded, specialized AI agents over technical knowledge.

Contextual AI is an enterprise platform for building retrieval-augmented AI agents that answer over a company's technical documentation, specifications, and institutional knowledge. It exposes individually tunable RAG components — document parsing, retrieval, reranking, and grounded generation — plus an agentic search and research layer, so teams in regulated, high-accuracy domains (finance, engineering, legal) can automate complex knowledge work while keeping answers grounded in source material. Available as multi-tenant SaaS, single-tenant SaaS, or private VPC, with developer APIs.

Retrieval engine with hybrid search, multi-vector, and custom ranking in one query.

TopK is a serverless retrieval engine that unifies vector (semantic), keyword (lexical), and multi-vector search with custom ranking in a single query and API — replacing the multi-database stack that RAG and search apps usually stitch together. Storage, inference (embedding, OCR, parsing), and queries run inside the engine or your own VPC, with SDKs for Python, JavaScript/TypeScript, and Rust plus a CLI and an MCP server.

Web-scale data extraction and a knowledge graph that grounds AI in facts.

Diffbot reads the public web like a person and turns it into structured data: an Extract API for articles and products, a Crawl service, a Natural Language API, and a Knowledge Graph of billions of entities and over a trillion facts. The graph is refreshed continuously, so AI systems can ground answers in current, verifiable data rather than model memory. Diffbot also ships its own factually grounded GraphRAG language model.

AI answer engine that synthesizes the web with cited sources.

Yahoo's AI-powered answer engine, synthesizing the open web, Yahoo data, and Yahoo content into concise responses with rich media, structured tables, and inline citations. Available as a standalone engine at scout.yahoo.com and inside the Yahoo Search app, with experiences woven across Yahoo Mail, News, Finance, and Sports. Built to drive traffic back to source publishers.

AI search and reading copilot that lives in your browser.

A browser extension that overlays an AI assistant on any web page: summarize long articles, ask questions about what you're reading, and use Focus to unpack complex terms in place. Its Cross-check feature pulls independent sources on a claim, and Suggestions surfaces related news, academic papers, and videos — so research happens without leaving the tab.

AI answer engine that searches the web and cites its sources.

Scira (formerly MiniPerplx) is an open-source, AI-powered answer engine — a Perplexity alternative built on the Vercel AI SDK. It plans a query, retrieves from the web, and returns concise answers with inline citations, drawing on 60+ models including Grok, Claude, Gemini, GPT, and DeepSeek. The AGPL-3.0 code is self-hostable, and a hosted app at scira.ai offers a free tier plus a Pro plan with connectors, memory, and voice.

AI enterprise search on a verified company knowledge layer.

Guru is an enterprise knowledge platform combining AI search, a verified company wiki, and an intranet, surfacing trusted answers inside the tools teams already use. Knowledge Agents give departments like HR, sales, and support cited responses grounded in governed content, backed by 100+ integrations and a browser extension.

AI universal search across your apps, files, and tools.

Dropbox Dash is an AI-powered universal search and knowledge workspace that indexes content across connected apps — Google Drive, OneDrive, Notion, Slack, and more — to return instant, sourced answers. It adds a unified start page, AI writing tools, and admin access controls, plus a business tier for teams.

Multimodal platform that makes images and video searchable and structured.

Coactive AI is an enterprise multimodal application platform that pulls context directly from the pixels and audio in images and video — no manual tagging or metadata required. Teams use it to semantically search, label, govern, and structure large visual libraries at scale, turning unstructured media into queryable data. It is aimed at media, retail, and other enterprises with vast image and video archives.

Analytics for how your brand shows up in AI search answers.

Peec AI is a generative engine optimization (GEO) analytics platform that tracks how brands appear across AI search engines like ChatGPT, Perplexity, and Google AI Overviews. It monitors mentions, sentiment, and the sources those engines cite, benchmarks competitors, and organizes prompts into reports so marketers can measure and improve AI-search visibility. The Berlin startup grew to 1,300+ customers and $4M+ ARR within its first year.

A discovery engine — a Wikipedia page for anything.

Globe Explorer is an AI discovery engine that turns any query into a single structured, visual page instead of a ranked list of links. It uses LLMs to break a topic into a hierarchy — broad overview down through subtopics — that you can explore, expand and share. It's free to use and aimed at learning and research rather than precise factual lookups.

Managed RAG-as-a-service with built-in hallucination control.

Vectara is an enterprise GenAI platform that delivers retrieval-augmented generation as a managed service, bundling ingestion, embedding, retrieval, reranking, and grounded generation behind one API. It ships first-party retrieval and generation models and a built-in hallucination-evaluation model (HHEM) to measure and reduce ungrounded answers. It targets regulated, accuracy-critical applications and AI agents.

AI-native vector search for multimodal product discovery.

Marqo is an end-to-end vector search engine that generates, stores, and retrieves embeddings for text and image data through a single API, so teams don't have to stitch together a separate embedding pipeline. It has since repositioned around AI-native ecommerce search, merchandising, and recommendations, with turnkey integrations for major commerce platforms.

Answer engine optimization — see and shape how AI search talks about your brand.

Profound monitors how brands show up in AI-generated answers across ChatGPT, Perplexity, Gemini, Claude, Copilot, Grok, and Google AI Overviews — tracking citations, share of voice, and sentiment. Its agents then act on the gaps, drafting briefs, articles, and optimization recommendations based on what the models actually cite. Customers include Target, Walmart, Ramp, and MongoDB.

Perplexity's AI browser with a sidebar assistant that acts across your tabs.

An AI-native Chromium browser from Perplexity that puts a persistent assistant in the sidebar — it summarizes pages, compares products, and takes actions like drafting emails or managing tabs using the context of what you're browsing. Perplexity search is the default, with Deep Research and voice mode built in. Free on Windows, macOS, Android, and iOS; an optional $5/month Comet Plus tier unlocks premium publisher content.

Multilingual AI search that turns answers into slides and mind maps.

An AI search and answer engine built around multilingual research — it searches across 25+ languages and synthesizes cited answers. Beyond text, Felo can generate editable presentations (AI PPT), mind maps, and agent workflows from a single query, making it a research-to-output tool rather than a pure search box.

Managed multimodal search over your text, PDFs, images, and video.

A fully managed search engine that indexes text, PDFs, tables, images, and video across 100+ languages without hand-tuning embeddings or a multi-stage pipeline. The Berlin team is best known for its open-source mxbai embedding and reranking models, which the hosted product builds on. Access it via dashboard, Python/TypeScript SDKs, or MCP.

AI search and research copilot with cited answers.

An AI search and research assistant that returns cited, fact-checked answers across the web, YouTube, and PDFs, with dedicated modes for academic literature review (Liner Scholar) and professional writing (Liner Write). It grew out of a popular web-highlighting tool and now spans browser, mobile, desktop, and a browser extension. The free tier covers everyday search; Pro unlocks more models, file uploads, and unlimited copilot use.

High-accuracy web search and research APIs for AI agents.

Parallel gives AI agents programmatic access to the web through a suite of APIs — a Search API, a Task API for structured extraction, and a Deep Research API for multi-hop questions — all served from its own proprietary web index and retrieval stack. It optimizes for high-signal, low-noise context fed straight into a model rather than ranking URLs for human clicks. Pricing is per request (the Search API starts at $0.005 per call) with a free allotment to start.

Open-source AI answer engine with a generative UI.

A self-hostable answer engine that returns comprehensive, cited answers with a generative UI that adapts per query rather than a list of links. It offers a Quick mode and an Adaptive multi-step research mode, saves searchable history, and is provider-agnostic across LLM and search backends. Apache-2.0, deployable to Vercel in one click, with a free hosted instance at morphic.sh.

Multimodal platform to search, reason over, and clip large volumes of video.

Reka Vision is an enterprise multimodal system that indexes large image and video libraries so teams can search by meaning, ask timestamp-aware questions, and auto-generate highlights and clips. It is built by Reka, a frontier multimodal-model lab, and is available via API, an MCP server, or a hosted app. Access is sales-led (request a demo).

Production-grade web search API for AI agents.

A web search API built for LLMs and agents. Linkup returns sourced, cited answers with full-text snippets in seconds, plus a Fetch endpoint for URL extraction and a Research endpoint that runs multi-step, chain-of-thought deep research. It integrates via MCP and SDK wrappers for OpenAI, LangChain, CrewAI, and more, and offers zero-data-retention enterprise hosting.

Search-foundation APIs — Reader, embeddings, and reranker — for grounding LLMs.

A suite of search-foundation APIs for retrieval and RAG: a Reader that turns any URL or web search into LLM-ready markdown, multilingual multimodal embeddings, and a reranker. One key spans every service, the Reader is open source, and the embedding models are also released as open weights for self-hosting.

Open-source document intelligence API for RAG-ready data.

A document parsing and intelligence API that turns complex PDFs, slides, Word docs, and images into clean, LLM/RAG-ready chunks. Chunkr runs layout analysis, OCR, reading-order detection, semantic chunking, and schema-based extraction, emitting HTML, Markdown, or JSON. Self-host the open-source pipeline or call the managed cloud API, which includes a free tier of 200 pages with no card required.

Paid, ad-free search engine with a built-in multi-model AI assistant.

Kagi is a subscription-based, ad-free search engine that puts result quality and privacy ahead of advertising. Every paid plan now includes Kagi Assistant, which lets you chat with leading models from OpenAI, Anthropic, Google, Mistral, and others against a fair-use token budget. Kagi also ships features like the Universal Summarizer and customizable result ranking.

Video intelligence API: search, classify, and summarize video.

Video understanding platform built on its own multimodal foundation models — Marengo for embeddings and semantic search, Pegasus for generative tasks like summaries and captions. Developers index video once and run natural-language search, classification, and analysis via API. Free tier with usage-based pricing beyond it.

Multi-model AI search and enterprise research agents.

You.com is an AI search and answers product offering multi-model chat over OpenAI, Anthropic, and Google models on web and mobile. It has shifted toward enterprise with ARI, a research agent that synthesizes cited reports across hundreds of sources, alongside Search and Research APIs for developers.

Open-source, self-hosted AI chat and enterprise search over your own docs.

Onyx (formerly Danswer) is an open-source AI chat and RAG platform that connects to your company's docs and apps for grounded, cited answers, and works with any LLM. It is self-hosted via Docker/Kubernetes and supports local models, keeping data on your own infrastructure. The core is MIT-licensed and free; an open-core model puts optional enterprise features under a separate license, and the vendor also offers a managed cloud.

All-in-one private AI app for chatting with your documents, with agents.

An all-in-one application for private, ChatGPT-style chat over your own documents, with built-in RAG, AI agents, and multi-user workspaces. Runs as a local desktop app (Mac/Windows/Linux) or self-hosted via Docker, and supports 40+ LLM providers plus local models with your own keys. Open source under MIT; Mintplex Labs also offers a paid hosted instance, making it freemium.

Autonomous AI agent that runs deep multi-source web research and writes cited reports.

An open-source autonomous research agent that plans a task, runs parallel multi-source web searches, validates sources, and synthesizes a cited report. It runs locally as a Python package, FastAPI server, or MCP server, and works with any LLM provider plus a search/retriever backend. Free and Apache-2.0 licensed — you only pay your own LLM and search-API costs.

Turn any website into clean, LLM-ready data — scrape, crawl, search.

A web data API for AI — scrape, crawl, map, and search pages into clean markdown or structured JSON, handling proxies, anti-bot, and JS rendering for you. Open-source core (AGPL) plus a hosted service; a default web-ingestion layer for agents and RAG pipelines.

Web search API built for LLM agents and RAG pipelines.

Search-as-a-tool for LLM agents — returns scrape-friendly results tuned for retrieval rather than ranking. Native integrations across LangChain, LangGraph, CrewAI, and the major agent surfaces.

AI search engine that answers with inline citations.

Hybrid search engine that returns synthesized answers with inline citations rather than a list of links. Ships a consumer product and the Sonar API for adding cited search to your own apps and agents.

Neural search API. Find pages by meaning, not keywords.

Semantic search engine that indexes the open web with embeddings — pass a description, get matching pages. Strong for research-style queries and find-similar workflows; formerly known as Metaphor.

Independent search index. Privacy-respecting, non-Google.

API access to Brave's independently-crawled search index. Useful when you need a non-Google source, want to dodge upstream rate limits, or have privacy posture requirements that rule out scraping competitors.