[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fD95647zpMPspJL7xZQpdudW9GEawHmHGgePLb3chVmU":3,"tool-freellm":16},{"siteName":4,"siteDescription":5,"siteUrl":6,"contactEmail":7,"socialLinks":8,"analyticsCode":10,"seo":11},"AllAI App","Call 500+ AI models through one compliant API KEY. Discover 35,000+ AI tools across 21 categories. Compare, test all ai app in one platform.","https:\u002F\u002Fallaiapp.com","contact@allaiapp.com",{"twitter":9,"github":9,"discord":9,"linkedin":9,"youtube":9},"","\u003C!-- Google tag (gtag.js) -->\n\u003Cscript async src=\"https:\u002F\u002Fwww.googletagmanager.com\u002Fgtag\u002Fjs?id=G-3NJE4SY4LE\">\u003C\u002Fscript>\n\u003Cscript>\n  window.dataLayer = window.dataLayer || [];\n  function gtag(){dataLayer.push(arguments);}\n  gtag('js', new Date());\n\n  gtag('config', 'G-3NJE4SY4LE');\n\u003C\u002Fscript>\n\u003C!-- Baidu Tongji (hm.js) -->\n\u003Cscript>\nvar _hmt = _hmt || [];\n(function() {\n  var hm = document.createElement(\"script\");\n  hm.src = \"https:\u002F\u002Fhm.baidu.com\u002Fhm.js?ee51358c8d4491dea1125cfc1d47a169\";\n  var s = document.getElementsByTagName(\"script\")[0];\n  s.parentNode.insertBefore(hm, s);\n})();\n\u003C\u002Fscript>",{"titleTemplate":12,"metaKeywords":13,"ogImage":14,"twitterSite":9,"defaultRobots":15,"googleVerify":9,"bingVerify":9,"baiduVerify":9,"organizationLogo":14},"%s | AllAI App","AI KEY,AI API KEY,Deepseek Key,Chatgpt Key,OpenAi Key,AI tools, AI directory, AI API gateway, AI models, artificial intelligence,","https:\u002F\u002Fallaiapp.com\u002Fog-image.png","index, follow",{"tool":17,"related":112},{"slug":18,"name":19,"cnName":19,"cn_name":19,"url":20,"pricingUrl":20,"category":21,"subcategory":9,"categories":22,"pricing":23,"priceFrom":24,"accent":25,"platforms":26,"inputs":29,"outputs":30,"languages":31,"tags":33,"capabilities":37,"rating":40,"votes":24,"reviewCount":24,"saves":24,"foundedYear":41,"founded":41,"developer":19,"verified":42,"editorPick":42,"featured":42,"status":43,"addedAt":44,"lastUpdated":44,"updatedAt":44,"en":45,"zh":79},"freellm","FreeLLM","https:\u002F\u002Ffreellm.net","code",[21],"free",0,"#0EA5E9",[27,28],"Web","API",[],[],[32],"en",[34,28,35,36],"LLM","free models","providers",[38,39],"llm-api","model-comparison",4.5,2026,false,"active","2026-10-30",{"tagline":46,"description":47,"features":48,"pros":54,"cons":58,"useCases":61,"pricingTiers":65,"faq":72},"Free LLM API & free AI API directory — compare free models from Google, Groq, OpenRouter, NVIDIA and more.","FreeLLM is a directory of free LLM APIs and free AI APIs. It lists free models from Google, Groq, OpenRouter, NVIDIA and others, with comparison across rate limits, context sizes and capabilities, so developers can pick the right free tier for their project.",[49,50,51,52,53],"Directory of free LLM and AI APIs","Providers: Google, Groq, OpenRouter, NVIDIA and more","Compare rate limits and context sizes","Filter by capability and provider","Updated list of free tiers",[55,56,57],"Saves hunting across providers","Free tiers compared in one place","Useful for prototypes and side projects",[59,60],"Directory, not a proxy itself","Free limits change frequently",[62,63,64],"Find a free LLM API for prototypes","Compare free tiers across providers","Select models by context and rate limits",[66],{"name":67,"price":68,"period":69,"features":70},"Free","0","forever",[71],"Directory access",[73,76],{"q":74,"a":75},"What is FreeLLM?","A directory comparing free LLM and AI API offerings.",{"q":77,"a":78},"Which providers are listed?","Google, Groq, OpenRouter, NVIDIA and others.",{"tagline":80,"description":81,"features":82,"pros":88,"cons":92,"useCases":95,"pricingTiers":99,"faq":105},"免费 LLM API 目录——横向比较 Google、Groq、OpenRouter、NVIDIA 等厂商的免费模型。","FreeLLM 是免费 LLM API 和免费 AI API 的目录，收录 Google、Groq、OpenRouter、NVIDIA 等厂商的免费模型，对比速率限制、上下文长度与能力，帮助开发者挑选适合项目的免费档位。",[83,84,85,86,87],"免费 LLM\u002FAI API 目录","覆盖 Google、Groq、OpenRouter、NVIDIA 等","对比速率限制与上下文长度","按能力与厂商筛选","免费档位持续更新",[89,90,91],"省去逐家官网翻找","免费档位一屏对比","适合原型与业余项目",[93,94],"本身是目录而非代理","免费额度变化频繁",[96,97,98],"为原型找免费 LLM API","跨厂商对比免费档","按上下文和限流选模型",[100],{"name":101,"price":68,"period":102,"features":103},"免费","永久",[104],"目录访问",[106,109],{"q":107,"a":108},"FreeLLM 是什么？","横向比较免费 LLM\u002FAI API 的目录。",{"q":110,"a":111},"收录哪些厂商？","Google、Groq、OpenRouter、NVIDIA 等。",[113,149,173,206,234,263,286,308,331,352,372,391],{"slug":114,"name":115,"cnName":116,"category":21,"categories":117,"subcategory":9,"pricing":23,"url":119,"logo":9,"accent":120,"tags":121,"rating":24,"saves":24,"featured":42,"editorPick":129,"priceFrom":24,"reviewCount":24,"updatedAt":130,"verified":42,"platforms":131,"monthlyVisits":24,"inputs":135,"outputs":138,"en":142,"zh":145,"similarity":148},"crawl4ai","Crawl4AI","Crawl4AI 大模型爬虫",[21,118],"research","https:\u002F\u002Fgithub.com\u002Funclecode\u002Fcrawl4ai","#22C55E",[122,34,123,124,125,126,127,128],"crawler","async","markdown","RAG","no api key","开源爬虫","大模型",true,"2026-09-13",[27,132,133,134],"CLI","Self-hosted","Python",[136,137],"URL","Text",[139,140,141],"Data","Markdown","Code",{"tagline":143,"description":144},"A free, open-source crawler built for LLMs: async, extremely fast, turns any site into clean markdown or structured JSON locally, with no API key or account.","Crawl4AI is an open-source Python crawler created specifically to feed large language models. It drives a browser asynchronously to render JavaScript-heavy pages and converts them into clean, LLM-friendly markdown or structured JSON, with configurable extraction strategies, chunking, metadata and screenshots. Because it runs on your own machine there is no API key, no account and no per-page fee — a point its author famously made after being frustrated by paid scraping services. It integrates with LangChain and similar frameworks and is optimized for high-throughput, cost-free local pipelines.",{"tagline":146,"description":147},"免费开源、为大模型而生的爬虫：异步极速，本地把任意网站转成干净 Markdown 或结构化 JSON，无需 API Key、无需账户。","Crawl4AI 是专为喂养大模型而做的开源 Python 爬虫。它异步驱动浏览器渲染 JS 重页面，转成干净、对 LLM 友好的 Markdown 或结构化 JSON，并提供可配置的抽取策略、分块、元数据与截图。因为完全在本机运行，所以无需 API Key、账户或按页付费——作者正是被付费爬虫服务激怒后几天内做出了它。它可与 LangChain 等框架集成，面向高吞吐、零成本的本地数据管线优化。",33,{"slug":150,"name":151,"cnName":152,"category":21,"categories":153,"subcategory":9,"pricing":23,"url":155,"logo":9,"accent":156,"tags":157,"rating":24,"saves":24,"featured":42,"editorPick":42,"priceFrom":24,"reviewCount":24,"updatedAt":130,"verified":42,"platforms":164,"monthlyVisits":24,"inputs":165,"outputs":166,"en":167,"zh":170,"similarity":148},"markitdown","MarkItDown","微软 MarkItDown 文档转Markdown",[21,154],"productivity","https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fmarkitdown","#0F6CBD",[124,158,159,160,161,125,34,162,163],"document conversion","pdf","office","ocr","格式转换","微软",[132,134,133],[136,137],[139,140,141],{"tagline":168,"description":169},"Microsoft's open-source tool that converts almost anything — PDF, Office docs, images (OCR), HTML, audio, archives and web pages — into clean Markdown for LLMs.","MarkItDown is a lightweight Microsoft open-source library\u002FCLI that transforms a very wide range of files into uniform Markdown so they can be fed to language models: Word, PowerPoint, Excel, PDF, HTML, CSV\u002FJSON\u002FXML, images (with OCR\u002Fdescriptions), audio (with transcription), EPUB, ZIP archives and more, plus web-page URLs. It optionally embeds metadata and image descriptions, is pip-installable and scriptable in Python, and has become a common first stage in RAG and document-AI pipelines where messy source formats must become clean text.",{"tagline":171,"description":172},"微软开源工具，把几乎一切——PDF、Office文档、图片(OCR)、HTML、音频、压缩包和网页——统一转成干净 Markdown 喂给大模型。","MarkItDown 是微软开源的轻量库\u002F命令行工具，能把极多种文件统一转成 Markdown 以便喂给语言模型：Word、PowerPoint、Excel、PDF、HTML、CSV\u002FJSON\u002FXML、图片（OCR\u002F描述）、音频（转写）、EPUB、ZIP 压缩包乃至网页URL。它可选择性保留元数据与图片描述，pip 即可安装、可用 Python 调用，已成为 RAG 与文档AI管线中“把杂乱源格式变成干净文本”的常见第一步。",{"slug":174,"name":175,"category":21,"categories":176,"subcategory":177,"pricing":23,"url":178,"logo":9,"accent":179,"tags":180,"rating":183,"saves":184,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":185,"monthlyVisits":9,"inputs":191,"outputs":192,"siteHealth":193,"en":199,"zh":202,"similarity":205},"github-32","PGVector",[],"dev-platforms","https:\u002F\u002Fgithub.com\u002Fpgvector\u002Fpgvector","#9C27B0",[181,182],"AI Code","Developer Tools",4.1,9918,[27,186,187,188,189,190],"Windows","Mac","Linux","iOS","Android",[137,141],[141,137],{"status":194,"http":195,"latencyMs":196,"finalUrl":178,"checkedAt":197,"title":198,"lang":32},"alive",200,664,"2026-09-29T09:45:15Z","GitHub - pgvector\u002Fpgvector: Open-source vector similarity search for Postgres",{"tagline":200,"description":201},"Postgres vector extension.","Open-source vector similarity search for Postgres. Contribute to pgvector\u002Fpgvector development by creating an account on GitHub.",{"tagline":203,"description":204},"AI驱动的code工具","PGVector 是一个 PostgreSQL 扩展，用于存储和查询向量数据。它支持精确和近似最近邻搜索，并集成了多种索引类型，如 HNSW 和 IVFFlat。该扩展易于安装，无需额外服务，可直接在现有 PostgreSQL 数据库中使用。适用于推荐系统、图像检索和自然语言处理等场景。",30,{"slug":207,"name":208,"category":21,"categories":209,"subcategory":210,"pricing":23,"url":211,"logo":9,"accent":212,"tags":213,"rating":215,"saves":216,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":217,"monthlyVisits":9,"inputs":218,"outputs":219,"siteHealth":220,"en":230,"zh":232,"similarity":205},"code-genius","Code Genius",[],"code-assistant","https:\u002F\u002Fwww.code-genius.dev","#00E676",[181,182,214],"Trending",4.7,9905,[27,28],[137,141],[141,137],{"status":221,"http":24,"latencyMs":24,"finalUrl":211,"checkedAt":222,"attempts":223},"error","2026-09-29T09:45:18Z",[224,227,228,229],[225,226],"http",301,[225,226],[225,226],[225,226],{"tagline":231,"description":231},"Streamlines React, Vue JS, and Tailwind CSS development.",{"tagline":203,"description":233},"Code Genius 是一款 AI 驱动的代码工具，专注于简化 React、Vue.js 和 Tailwind CSS 的开发流程。它提供智能代码补全、实时错误检测和代码解释功能，支持 Python、JavaScript、TypeScript 等多种编程语言，帮助开发者提高编码效率，快速理解和修复代码问题。目前免费使用，是开发者的得力助手。",{"slug":235,"name":236,"category":21,"categories":237,"subcategory":238,"pricing":23,"url":239,"logo":9,"accent":240,"tags":241,"rating":244,"saves":245,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":246,"monthlyVisits":247,"inputs":248,"outputs":251,"siteHealth":252,"en":257,"zh":260,"similarity":205},"langflow","LangFlow",[154],"ai-agents","https:\u002F\u002Fwww.langflow.org","#00BCD4",[242,243],"AI Agent","Automation",4.3,9855,[27],199498,[137,249,250],"Image","Audio",[137,249,141],{"status":194,"http":195,"latencyMs":253,"finalUrl":254,"checkedAt":255,"title":256,"lang":32},69,"https:\u002F\u002Fwww.langflow.org\u002F","2026-09-29T10:15:07Z","Langflow | Low-code AI builder for agentic and RAG applications",{"tagline":258,"description":259},"Visual LangChain builder.","Langflow is a low-code AI builder for agentic and retrieval-augmented generation (RAG) apps. Code in Python and use any LLM or vector database.",{"tagline":261,"description":262},"AI驱动的agent工具","LangFlow 是一款基于 LangChain 的可视化构建工具，旨在通过直观的拖拽式界面简化 AI 工作流的创建与管理。它支持自主 AI 代理，能够独立完成任务，并允许用户设计自定义自动化流程。同时，LangFlow 可连接外部 API 和服务，扩展功能边界，并提供监控功能，帮助用户跟踪代理性能和任务进度。该工具免费使用，适合希望快速搭建和部署 AI 应用的开发者。",{"slug":264,"name":265,"category":21,"categories":266,"subcategory":238,"pricing":23,"url":268,"logo":9,"accent":269,"tags":270,"rating":40,"saves":271,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":272,"monthlyVisits":273,"inputs":274,"outputs":275,"siteHealth":276,"en":281,"zh":284,"similarity":205},"donely","Donely",[267],"business","https:\u002F\u002Fdonely.ai","#F7931E",[242,243],9822,[27],9641,[137,249,250],[137,249,141],{"status":194,"http":195,"latencyMs":277,"finalUrl":278,"checkedAt":279,"title":280,"lang":32},1327,"https:\u002F\u002Fdonely.ai\u002F","2026-09-29T10:15:15Z","Autonomous Security Agent - Continuous, Exploit-Validated Pentesting",{"tagline":282,"description":283},"Run OpenClaw employees fleetwide without managing servers.","An AI employee that pentests like an adversary, validates like an auditor, and reports like an engineer. Proof, not probability.",{"tagline":261,"description":285},"Donely 是一款 AI 驱动的智能体工具，专注于让用户无需管理服务器即可在团队范围内运行 OpenClaw 员工。它提供自主任务执行、可视化工作流设计、外部工具集成以及性能监控等功能，帮助团队轻松实现自动化。Donely 免费使用，适合希望快速部署 AI 智能体的个人和团队。",{"slug":287,"name":288,"category":21,"categories":289,"subcategory":238,"pricing":23,"url":290,"logo":9,"accent":179,"tags":291,"rating":293,"saves":294,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":295,"monthlyVisits":9,"inputs":296,"outputs":297,"siteHealth":298,"en":303,"zh":306,"similarity":205},"runtm","Runtime",[154],"https:\u002F\u002Fruntm.com",[242,243,292],"Popular",4.4,9813,[27],[137,249,250],[137,249,141],{"status":194,"http":195,"latencyMs":299,"finalUrl":300,"checkedAt":301,"title":302,"lang":32},163,"https:\u002F\u002Fwww.runtm.com\u002F","2026-09-29T10:15:17Z","Runtime - AI agents for financial operations",{"tagline":304,"description":305},"Centralize and govern coding agents shipping work.","AI agents for financial operations. Investigate payment issues, handle bank requests, and review merchants with scoped access, approvals, and an audit trail.",{"tagline":261,"description":307},"Runtime 是一个专注于集中管理和治理编码代理（coding agents）的工具。它提供自主任务执行、可视化工作流构建、外部工具集成以及监控功能，帮助团队高效协调 AI 代理的工作。该平台目前免费使用，适合希望自动化编码流程并提升开发效率的个人和团队。通过直观的界面，用户可以设计自定义自动化流程，并实时跟踪代理的进度和性能。",{"slug":309,"name":310,"category":21,"categories":311,"subcategory":177,"pricing":23,"url":312,"logo":9,"accent":313,"tags":314,"rating":316,"saves":317,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":318,"monthlyVisits":9,"inputs":319,"outputs":320,"siteHealth":321,"en":327,"zh":329,"similarity":205},"learn","Semantic Kernel",[],"https:\u002F\u002Flearn.microsoft.com\u002Fsemantic-kernel","#2196F3",[181,182,315],"Editor's Pick",4.8,9722,[27,28],[137,141],[141,137],{"status":194,"http":195,"latencyMs":322,"finalUrl":323,"checkedAt":324,"title":325,"lang":326},223,"https:\u002F\u002Flearn.microsoft.com\u002Fen-us\u002Fsemantic-kernel\u002F","2026-10-09T09:39:17Z","Semantic Kernel documentation","en-us",{"tagline":328,"description":328},"Microsoft's AI orchestration SDK.",{"tagline":203,"description":330},"Semantic Kernel 是微软推出的 AI 编排 SDK，旨在帮助开发者将大型语言模型集成到应用程序中。它支持多种编程语言，如 Python、JavaScript 和 TypeScript，并提供代码补全、错误检测和代码解释等功能，以提升开发效率。作为免费的开源工具，它适合希望在项目中引入 AI 能力的开发者。",{"slug":332,"name":333,"category":21,"categories":334,"subcategory":177,"pricing":23,"url":335,"logo":9,"accent":336,"tags":337,"rating":316,"saves":338,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":339,"monthlyVisits":340,"inputs":341,"outputs":342,"siteHealth":343,"en":348,"zh":350,"similarity":205},"warp","Warp",[],"https:\u002F\u002Fwww.warp.dev","#E040FB",[181,182],9719,[27,186,187,188],902504,[137,141],[141,137],{"status":194,"http":195,"latencyMs":344,"finalUrl":345,"checkedAt":346,"title":347,"lang":32},515,"https:\u002F\u002Fwww.warp.dev\u002F","2026-10-02T15:45:05Z","Warp — The Open Platform for Automating Development",{"tagline":349,"description":349},"Warp Terminal -AI编程工具：AI命令补全+错误智能诊断：Warp Terminal是一款用Rust编写的现代化、AI驱动的终端模拟器，提供AI命令补全、自然语言解释错误、块级编辑与团队协作，帮助开发者提升命令行效率。",{"tagline":203,"description":351},"Warp Code 是 Warp 团队最新推出的AI编程工具，构建从 Prompt 到生产环境的完整 AI 辅助编程闭环，具备顶级编码 Agent、代码审查、代码编辑和项目管理等功能，让开发者能在 Warp 内高效完成从编写 Prompt 到代码生成、审查、修改直至发布的全流程，显著提升开发效率和代码质量",{"slug":353,"name":354,"category":21,"categories":355,"subcategory":177,"pricing":23,"url":356,"logo":9,"accent":357,"tags":358,"rating":183,"saves":359,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":360,"monthlyVisits":9,"inputs":361,"outputs":362,"siteHealth":363,"en":367,"zh":370,"similarity":205},"github-30","TensorRT-LLM",[],"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FTensorRT-LLM","#4CAF50",[181,182],9646,[27,186,187,188,189,190],[137,141],[141,137],{"status":194,"http":195,"latencyMs":364,"finalUrl":356,"checkedAt":365,"title":366,"lang":32},620,"2026-09-29T10:45:12Z","GitHub - NVIDIA\u002FTensorRT-LLM: TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models",{"tagline":368,"description":369},"NVIDIA's LLM inference optimizer.","TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. Tensor...",{"tagline":203,"description":371},"TensorRT-LLM 是 NVIDIA 推出的 LLM 推理优化工具，专注于提升大型语言模型的推理性能。它通过先进的优化技术，加速模型在 NVIDIA GPU 上的运行，降低延迟并提高吞吐量。该工具适用于多种编程语言，提供代码补全、实时错误检测和代码解释等功能，帮助开发者更高效地编写和调试代码。TensorRT-LLM 完全免费，是开发者优化 LLM 推理的理想选择。",{"slug":373,"name":374,"category":21,"categories":375,"subcategory":376,"pricing":23,"url":377,"logo":9,"accent":378,"tags":379,"rating":215,"saves":382,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":383,"monthlyVisits":9,"inputs":384,"outputs":385,"en":386,"zh":388,"similarity":205},"trymirai","Mirai",[154],"general-chatbot","https:\u002F\u002Ftrymirai.com","#40C4FF",[380,381],"AI Chatbot","Conversational AI",9608,[27],[137,249,250],[137,249,141],{"tagline":387,"description":387},"Run AI models on-device for privacy and speed.",{"tagline":389,"description":390},"AI驱动的chatbot工具","Mirai 是一款运行在设备上的 AI 聊天机器人工具，旨在提供隐私保护和快速响应。它支持自然对话、多轮对话，并能根据用户自定义指令调整行为。Mirai 利用本地计算，减少数据传输，从而增强隐私安全。其知识库基于大量数据训练，确保回答准确。该工具免费使用，适合注重隐私和速度的用户。",{"slug":392,"name":393,"category":21,"categories":394,"subcategory":238,"pricing":23,"url":395,"logo":9,"accent":313,"tags":396,"rating":397,"saves":398,"editorPick":42,"priceFrom":67,"reviewCount":24,"updatedAt":9,"platforms":399,"monthlyVisits":9,"inputs":400,"outputs":401,"siteHealth":402,"en":407,"zh":410,"similarity":205},"docs-3","LlamaIndex Evaluators",[118],"https:\u002F\u002Fdocs.llamaindex.ai",[242,243],4.2,9572,[27],[137,249,250],[137,249,141],{"status":194,"http":195,"latencyMs":403,"finalUrl":404,"checkedAt":405,"title":406,"lang":32},128,"https:\u002F\u002Fdevelopers.llamaindex.ai\u002Fpython\u002Fframework\u002F","2026-10-09T09:39:19Z","LlamaIndex Framework, by the makers of LlamaParse",{"tagline":408,"description":409},"LlamaIndex learning.","LlamaIndex is the company behind LlamaParse, the document parsing platform for AI agents. The LlamaIndex framework is our original open-source toolkit for RAG and agents.",{"tagline":261,"description":411},"LlamaIndex Evaluators 是 LlamaIndex 提供的评估工具，专注于 AI 代理的自动化任务执行。它支持可视化工作流构建，可连接外部 API 和服务，并提供监控功能以跟踪代理性能和任务进度。该工具旨在帮助开发者更高效地构建和评估 AI 代理，适用于需要自动化复杂任务的场景。"]