openui

The next generation of ai apps won't reply with text. they'll reply with interfaces. Openui is an open-source language built for exactly that, using up to 67% fewer tokens than json 👇 thesysdev/openui

Generative UI has an open standard now. OpenUI (@openuidev) streams model-written UI into your own components. Productboard is an adopter. The model paints the screen. The question is whose design system it uses. thesysdev/openui #AgenticDesign #SortedPixels

@augustindebs @ChatGPT Oui mais celui qui se rapproche le plus c’est lui thesysdev/openui (j’avais fait un post dessus également)

Builds generative UI interfaces using a compact, streaming-first language that consumes up to 67% fewer tokens than JSON thesysdev/openui t.co

🚀 npx @openuidev/cli@latest create scaffolds a streaming chat app. An Agent Skill helps Claude Code, Codex, Cursor & Copilot build with it. #React #Vue #Svelte #Angular #LangChain thesysdev/openui

Top 10 Trending AI GitHub Tools (#1 Harvey Cold-Emails Strangers for You) A countdown of 10 real, currently trending open-source AI GitHub projects. This week agents got everything a new hire gets on day one: memory, a map, a style guide, a supervisor. The #1 pick finds prospects, writes cold emails and books meetings on its own. In countdown order: GuppyLM, a 9M-parameter model that trains in five minutes and thinks it is a fish. Impeccable, a design language with 62 detector rules for AI-built frontends. OpenUI, an open standard for generative UI that streams interfaces instead of paragraphs. billion-context, a proxy that folds and unfolds context for month-long coding sessions. ai-memory, shared long-term memory across 20+ coding agents. Kronos, a foundation model for financial candlestick charts, small enough for a normal GPU. TokenSpeed, an inference engine for agent workloads on Blackwell-class hardware. Graphify, which turns a codebase into a queryable knowledge graph, offline. Forge, guardrails that make small local models reliable at tool calling. And at #1, Harvey, an autonomous sales agent with an approval outbox you should keep switched on. Projects in this video: GuppyLM: arman-bd/guppylm Impeccable: pbakaus/impeccable OpenUI: thesysdev/openui billion-context: ranxianglei/billion-context ai-memory: akitaonrails/ai-memory Kronos: shiyu-coder/Kronos TokenSpeed: lightseekorg/tokenspeed Graphify: Graphify-Labs/graphify Forge: antoinezambelli/forge Harvey: ethanplusai/harvey 0:00 Intro 0:43 #10 GuppyLM 1:55 #9 Impeccable 3:15 #8 OpenUI 4:30 #7 billion-context 5:48 #6 ai-memory 7:07 #5 Kronos 8:27 #4 TokenSpeed 9:48 #3 Graphify 11:07 #2 Forge 12:24 #1 Harvey 13:35 Outro Contact: digestibleinfohub@gmail.com #Harvey #AI #GitHub #OpenSource #DevTools

4,774 Stars Ek Hafte Mein, Aur Ye Sirf Number One Hai: Is Hafte Ke 5 GitHub Repos #Shorts Top 5 trending GitHub repos, Hindi mein - har number aaj check kiya gaya hai. Is week ke top trending repos, 201k+ stars combined: 1. anthropics/claude-plugins-official (44k stars) - Official, Anthropic-managed directory of high quality Claude Code Plugins. 2. Alishahryar1/free-claude-code (62k stars) - Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly) 3. hmjz100/LinkSwift (27k stars) - 一个基于 JavaScript 的网盘文件下载地址获取工具。基于【网盘直链下载助手】修改 ，支持 百度网盘 / 阿里云盘 / 中国移动云盘 / 天翼云盘 / 迅雷云盘 / 夸克网盘 / UC网盘 / 123云盘 八大网盘 4. HKUDS/CLI-Anything (56k stars) - "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: clianything.cc 5. thesysdev/openui (13k stars) - The Open Standard for Generative UI Repo links: anthropics/claude-plugins-official Alishahryar1/free-claude-code hmjz100/LinkSwift HKUDS/CLI-Anything thesysdev/openui Comment karo "LIVE" - saare links DM mein bhej dunga. Kisi dev dost ko bhejo jise ye chahiye. Save karlo, baad mein kaam aayega. Naman AI - AI aur dev tools, Hindi mein. Har din naya repo, har number verified. #Shorts #GitHub #AI #OpenSource #TechHindi [github trending repos, best github repositories 2026, top 5 github repos, open source projects for beginners, ai coding agents, github stars] [coding in hindi, tech shorts hindi, programming for beginners, developer tools 2026, ai tools for developers, learn coding free] [bn] এক সপ্তাহে ৪,৭৭৪ স্টার, আর এটি কেবল প্রথমটি: এই সপ্তাহের ৫টি গিটহাব রিপো #Shorts হিন্দিতে শীর্ষ ৫টি ট্রেন্ডিং গিটহাব রিপো - প্রতিটি নম্বর আজ চেক করা হয়েছে। এই সপ্তাহের শীর্ষ ট্রেন্ডিং রিপো, সম্মিলিতভাবে ২০১ হাজারেরও বেশি স্টার: ১. anthropics/claude-plugins-official (৪৪ হাজার স্টার) - উচ্চ মানের ক্লড কোড প্লাগইনগুলির অফিসিয়াল, অ্যানথ্রোপিক-পরিচালিত ডিরেক্টরি। ২. Alishahryar1/free-claude-code (৬২ হাজার স্টার) - আপনার টার্মিনাল, অ্যাপ, IDE, বা ফোন থেকে বিনামূল্যে ক্লড কোড, কোডেক্স, ভিএসকোড, পাই, এবং ওপেনকোড (এবং আরও ৬টি হারনেস) ব্যবহার করুন (১.৩ বিলিয়নেরও বেশি ফ্রি টোকেন), এবং এখন ওপেনক্ল-এর মতো নেটিভ ব্রাউজার সেশনের মাধ্যমে ব্রাউজার থেকেও ব্যবহার করুন (মাল্টি-হারনেস + মাল্টি-মডেল) (ভয়েস সাপোর্টেড + ToS ফ্রেন্ডলি)। ৩. hmjz100/LinkSwift (২৭ হাজার স্টার) - নেটওয়ার্ক ডিস্ক ফাইলের ডাউনলোড অ্যাড্রেস পাওয়ার জন্য একটি জাভাস্ক্রিপ্ট-ভিত্তিক টুল। সংশোধিত ‘ডাইরেক্ট ডাউনলোড অ্যাসিস্ট্যান্ট ফর ক্লাউড ড্রাইভস’-এর উপর ভিত্তি করে, এই সংস্করণটি আটটি প্রধান ক্লাউড স্টোরেজ পরিষেবা সমর্থন করে: বাইডু ক্লাউড ড্রাইভ, আলিবাবা ক্লাউড ড্রাইভ, চায়না মোবাইল ক্লাউড ড্রাইভ, তিয়ানয়ি ক্লাউড ড্রাইভ, জুনলেই ক্লাউড ড্রাইভ, কোয়ার্ক ক্লাউড ড্রাইভ, ইউসি ক্লাউড ড্রাইভ, এবং ১২৩ ক্লাউড ড্রাইভ। 4. HKUDS/CLI-Anything (56k stars) - "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: clianything.cc 5. thesysdev/openui (13k তারা) - জেনারেটিভ UI এর জন্য ওপেন স্ট্যান্ডার্ড রেপো লিঙ্ক: anthropics/claude-plugins-official Alishahryar1/free-claude-code hmjz100/LinkSwift HKUDS/CLI-%E0%A6%AF%E0%A7%87%E0%A6%95%E0%A7%8B%E0%A6%A8%E0%A7%8B কিছু thesysdev/openui মন্তব্য করুন "লাইভ" - সারা লিঙ্ক ডিএম মে ভেজ দুঙ্গা। কিসি দেব দোস্ত কো ভেজো জিসে ইয়ে চাহিয়ে। করলো বাঁচাও, বাদ মে কাম আয়েগা। নমন এআই - এআই এবং ডেভ টুলস, হিন্দিতে। প্রতিদিন নতুন রিপো, প্রতিটি নম্বর ভেরিফাইড। #শর্টস #গিটহাব #এআই #ওপেনসোর্স #টেকহিন্দি [গিটহাব ট্রেন্ডিং রিপো, সেরা গিটহাব রিপোজিটরি ২০২৬, সেরা ৫টি গিটহাব রিপো, নতুনদের জন্য ওপেন সোর্স প্রোজেক্ট, এআই কোডিং এজেন্ট, গিটহাব স্টারস] [হিন্দিতে কোডিং, টেক শর্টস হিন্দি, নতুনদের জন্য প্রোগ্রামিং, ডেভেলপার টুলস ২০২৬, ডেভেলপারদের জন্য এআই টুলস, বিনামূল্যে কোডিং শিখুন] [en-US] 4,774 Stars in a Week, and This Is Only Number One: 5 GitHub Repos of This Week #Shorts Top 5 trending GitHub repos, Hindi mein - har number aaj check kiya gaya hai. Is this week top trending repos, 201k+ stars combined: 1. anthropics/claude-plugins-official (44k stars) - Official, Anthropic-managed directory of high quality Claude Code Plugins. 2. Alishahryar1/free-claude-code (62k stars) - Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly) 3. hmjz100/LinkSwift (27k stars) - A JavaScript-based tool for obtaining the download address of network disk files. Based on the modified "Direct Download Assistant for Cloud Drives", this version supports eight major cloud storage services: Baidu Cloud Drive, Alibaba Cloud Drive, China Mobile Cloud Drive, Tianyi Cloud Drive, Xunlei Cloud Drive, Quark Cloud Drive, UC Cloud Drive, and 123 Cloud Drive. 4. HKUDS/CLI-Anything (56k stars) - "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: clianything.cc 5. thesysdev/openui (13k stars) - The Open Standard for Generative UI Repo links: anthropics/claude-plugins-official Alishahryar1/free-claude-code hmjz100/LinkSwift HKUDS/CLI-Anything thesysdev/openui Comment karo "LIVE" - saare links DM mein bhej dunga. Kisi dev dost ko bhejo jise ye chahiye. Save karlo, baad mein kaam aayega. Naman AI - AI aur dev tools, Hindi mein. Har din naya repo, har number verified. #Shorts #GitHub #AI #OpenSource #TechHindi [github trending repos, best github repositories 2026, top 5 github repos, open source projects for beginners, ai coding agents, github stars] [coding in hindi, tech shorts hindi, programming for beginners, developer tools 2026, ai tools for developers, learn coding free]

𝗥𝗲𝗽𝗼: 𝗢𝗽𝗲𝗻𝗨𝗜 🤖 🔗 thesysdev/openui An open-source framework for building AI-generated user interfaces with React. Create interactive components like charts, tables, and forms dynamically instead of relying on text-only AI responses.

OpenUI: The Open Standard for Generative UI (12k stars) What it does All text below is quoted directly from the project README. >OpenUI is a full-stack, framework-agnostic Generative UI framework built around a compact, streaming-first language. A framework-agnostic core powers first-party runtimes for React, Vue, Svelte, and Angular, and React additionally ships built-in component libraries and ready-to-use chat interfaces. OpenUI Lang uses up to 67% fewer tokens than JSON. Core capabilities: - OpenUI Lang - A compact language for structured UI generation designed for streaming output. - Built-in component libraries - Charts, forms, tables, layouts, and more, ready to use or extend. - Prompt generation from your component library - Generate model instructions directly from the components you allow. - Streaming renderer - Parse and render model output progressively in React, Vue, Svelte, or Angular as tokens arrive. - Chat and app surfaces - Use the same foundation for assistants, copilots, and broader interactive product flows. >npx @openuidev/cli@latest create --name genui-chat-app Repository thesysdev/openui Repository stats |Metric|Value| |:-|:-| |Stars|11,623| |Forks|791| |Last release|@openuidev/cli@0.6.0 (2026-10-15)| |Languages|TypeScript 90.6%, SCSS 8%| |Contributors|58| |AI contributor accounts (Claude / Codex / Cursor / Copilot)|Copilot: 1, cursoragent: 17, claude: 109| |AI co-authored commits|Claude: 113, Cursor: 18| Main contributor |Field|Value| |:-|:-| |Name|Subham Das| |GitHub|@i-subham23| |Commits to this repo|164 (top contributor)| |Bio|Hello World!| |Company / Location|Thesys| |Followers|0| |Public repos|5| |On GitHub since|2024-06-17|

how good are small local LLMs at generating interfaces instead of text? with chatgpt recently launching intelligent ui, i've been looking into the different approaches to generative ui and wondering how well smaller local models would actually handle this. for anyone unfamiliar with the concept, the idea is pretty straightforward. instead of an llm responding with a wall of text or markdown, it generates actual interface components like charts, tables, forms, and interactive elements depending on what the user asks. there are a few projects working on this, like vercel's json-render, google's a2ui, and openui (which coincidentally just launched its 1.0 version). they have slightly different approaches to how the interfaces are generated, but the underlying idea is similar. what i'm particularly curious about is where smaller models start falling apart. getting an llm to produce valid json or some other structured format is one thing, but generating a useful interface feels like a different problem altogether. say you ask a model to analyze your monthly expenses. it needs to figure out that a chart would be more useful than a table for showing spending trends, decide which data belongs on which axis, and maybe include a few filters to let you explore the results. and all of that needs to happen without hallucinating data, picking the wrong components, or generating an interface that technically renders but makes absolutely no sense. i've been looking into how different projects tackle this. json-render and a2ui rely on structured json-based representations, while openui uses its own DSL called OpenUI Lang, which is specifically designed for generating interfaces with llms. according to OpenUI's benchmarks, their approach uses around 67% fewer tokens than comparable json-based formats. which makes me wonder whether reducing the amount of structure a model needs to generate would make a meaningful difference for smaller local models, especially on consumer hardware. another interesting thing i came across is that the OpenUI team released OUI-1, a dedicated open-weight model for generative ui based on diffusiongemma, with 4b active parameters. they report a 71.7% score on their generative ui benchmark, although i'd take any project-maintained benchmark with a grain of salt until there's more independent testing. but it raises an interesting question: would a smaller model specifically trained to generate interfaces outperform a much larger general-purpose chat model on this particular task? and if so, would it make more sense to have one model handle the actual reasoning and tool calls, while a smaller dedicated model handles turning the results into an interface? for example, you could have your main agent analyze a dataset and return structured information, then pass that to a smaller locally running model whose only job is deciding how that information should be presented. i wonder how much this would reduce latency, and whether the additional model call would even be worth it. the other question is what would actually be the best way to evaluate this. most benchmarks i've seen focus on whether the generated output is valid and renders successfully. but a model could generate a perfectly valid chart that represents the data incorrectly, or choose a table when a simple form would have been far more useful. so how do you meaningfully measure whether a generated interface is actually good? some links for anyone interested: - OpenUI 1.0: thesysdev/openui - OUI-1 (4B active parameters): huggingface.co - Generative UI Bench: openui.com - json-render: vercel-labs/json-render - A2UI: a2ui-project/a2ui curious if anyone here has experimented with this using local models. if you were testing generative ui on consumer hardware, where would you start? forms, tables, charts, or something more complex? and would you rather use your existing chat model for everything, or have a separate smaller model dedicated to generating the interface? i'd be interested in hearing what models people have tried and where they started running into limitations.