LogoThread Easy
  • Explorar
  • Criar thread
LogoThread Easy

Seu parceiro completo para threads do Twitter

© 2025 Thread Easy All Rights Reserved.

Explorar

Newest first — browse tweet threads

Keep on to blur preview images; turn off to show them clearly

incredibly based. Now that we have fast long contexts and an abundance of compute, it's about damn time to explore byte-level models again. Meta has disappointed me with MegaByte, but Meta generally is bad at execution. This path is not yet closed…

incredibly based. Now that we have fast long contexts and an abundance of compute, it's about damn time to explore byte-level models again. Meta has disappointed me with MegaByte, but Meta generally is bad at execution. This path is not yet closed…

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

avatar for Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Tue Dec 16 01:12:03
RT @indie_maker_fox: 2025年,运营了一整年的 Discord

人数达到 2700+,发送消息 5200+

解答了无数的问题,分享了无数个经验

https://t.co/nHJuQKgx3H

免费入群,聊天记录全公开,build in public…

RT @indie_maker_fox: 2025年,运营了一整年的 Discord 人数达到 2700+,发送消息 5200+ 解答了无数的问题,分享了无数个经验 https://t.co/nHJuQKgx3H 免费入群,聊天记录全公开,build in public…

🔥 The best AI SaaS boilerplate - https://t.co/VyNtTs0jSX 🚀 The best directory boilerplate with AI - https://t.co/wEvJ1Dd8aR 🎉 https://t.co/bh1RxeERuY & https://t.co/zubXJCoY92 & https://t.co/tfQf8T7gGF

avatar for Fox@MkSaaS.com
Fox@MkSaaS.com
Tue Dec 16 01:10:14
RT @PGelsinger: Excited to launch the Flourishing AI Christian (FAI-C) Benchmark today. This builds on the original Flourishing AI Benchmar…

RT @PGelsinger: Excited to launch the Flourishing AI Christian (FAI-C) Benchmark today. This builds on the original Flourishing AI Benchmar…

Author. Coder. CTO. θηριομάχης. Building: https://t.co/otXT4Wy6WR. Writing: https://t.co/dBPBtyCIHw.

avatar for Jon Stokes
Jon Stokes
Tue Dec 16 01:09:28
you can buy anything on Aliexpress
Jack Ma's contribution to humanity is underappreciated
there are massive spillover benefits too, with Alibaba machine funding Hangzhou industrial policy and Alibaba alumni creating and supporting new startups

you can buy anything on Aliexpress Jack Ma's contribution to humanity is underappreciated there are massive spillover benefits too, with Alibaba machine funding Hangzhou industrial policy and Alibaba alumni creating and supporting new startups

We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1

avatar for Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
Tue Dec 16 01:09:03
[开源推荐] HyperBookLM: 带 web-agent 的开源版 NotebookLM,增加了对网页内容的智能抓取能力,来自 @hyperbrowser 团队

项目核心功能
项目允许用户聚合多种来源的信息,并通过 AI 进行深度分析和交互:
· 来源聚合:
  · 支持输入网页 URL:使用 Hyperbrowser SDK 实时抓取并提取网站内容
  · 支持上传 PDF 或 TXT 文件:在客户端直接解析和处理
· AI 分析与输出:
  · 生成综合研究摘要(包括关键洞见)
  · 创建交互式思维导图(React Flow)
  · 产生播客风格的音频概述(ElevenLabs 语音合成)
  · 自动生成演示幻灯片(Slides)
  · 交互式聊天:针对所有来源内容提问,支持流式响应
 
技术栈
· 前端框架:Next.js 15(App Router)、React 19
· UI 与样式:Tailwind CSS、shadcn/ui、Framer Motion
· 核心集成:
  · 网页抓取:Hyperbrowser SDK
  · AI 模型:
    · OpenAI:gpt-5-nano(聊天)、gpt-4o-mini(摘要、思维导图)。
    · Google Gemini:gemini-3-pro-image-preview(PPT)。
    · ElevenLabs:eleven_turbo_v2_5(音频)。
  · 可视化:React Flow(思维导图)。
  · PDF 处理:unpdf 库。

项目结构
· app/:核心页面和 API 路由(包括刮取、聊天、摘要、音频、思维导图、幻灯片等后端集成)
· components/:UI 组件,如聊天界面、思维导图、来源面板、输出面板
· lib/:API 客户端和类型定义

应用场景
· 内容重制:抓取博客生成 LinkedIn PPT 或播客
· 竞争分析:分析对手页面生成 SWOT 思维导图
· 研究聚合:整合白皮书和文档,形成可查询知识库

开源地址

[开源推荐] HyperBookLM: 带 web-agent 的开源版 NotebookLM,增加了对网页内容的智能抓取能力,来自 @hyperbrowser 团队 项目核心功能 项目允许用户聚合多种来源的信息,并通过 AI 进行深度分析和交互: · 来源聚合: · 支持输入网页 URL:使用 Hyperbrowser SDK 实时抓取并提取网站内容 · 支持上传 PDF 或 TXT 文件:在客户端直接解析和处理 · AI 分析与输出: · 生成综合研究摘要(包括关键洞见) · 创建交互式思维导图(React Flow) · 产生播客风格的音频概述(ElevenLabs 语音合成) · 自动生成演示幻灯片(Slides) · 交互式聊天:针对所有来源内容提问,支持流式响应 技术栈 · 前端框架:Next.js 15(App Router)、React 19 · UI 与样式:Tailwind CSS、shadcn/ui、Framer Motion · 核心集成: · 网页抓取:Hyperbrowser SDK · AI 模型: · OpenAI:gpt-5-nano(聊天)、gpt-4o-mini(摘要、思维导图)。 · Google Gemini:gemini-3-pro-image-preview(PPT)。 · ElevenLabs:eleven_turbo_v2_5(音频)。 · 可视化:React Flow(思维导图)。 · PDF 处理:unpdf 库。 项目结构 · app/:核心页面和 API 路由(包括刮取、聊天、摘要、音频、思维导图、幻灯片等后端集成) · components/:UI 组件,如聊天界面、思维导图、来源面板、输出面板 · lib/:API 客户端和类型定义 应用场景 · 内容重制:抓取博客生成 LinkedIn PPT 或播客 · 竞争分析:分析对手页面生成 SWOT 思维导图 · 研究聚合:整合白皮书和文档,形成可查询知识库 开源地址

邵猛,中年失业程序员 😂 专注 - Context Engineering, AI Agents. 分享 - AI papers, apps and OSS. ex Microsoft MVP 合作 - 私信/邮箱:shaomeng@outlook.com 📢 公众号/小红书: AI 启蒙小伙伴

avatar for meng shao
meng shao
Tue Dec 16 01:06:09
Best talk on the past/present/future of coding evals at @aidotengineer CODE. basically track it by horizons:

seconds: Copilot Arena
minutes: LiveCodeBench
multiple mins: RepoChat
~hour: GSO
multiple hours: Syzygy
days: ???

https://t.co/exTyCakFxe

when I invited @StringChaos to speak I wasn't even aware that he had joined @Cursor_ai fulltime; I was just hoping for the story of LiveCodeBench. Instead we got more than I ever dreamed of; a comprehensive look at all the ways in which koding evals have evolved in the last 3 years and everything that the community has learned, including o3's reward hacking, and now benchmarks to migrate REALLY large amounts of code and online evals to adjust for latency and real world constraints. ends with some actually prescriptive recommendations if you're also building coding evals of your own.

Best talk on the past/present/future of coding evals at @aidotengineer CODE. basically track it by horizons: seconds: Copilot Arena minutes: LiveCodeBench multiple mins: RepoChat ~hour: GSO multiple hours: Syzygy days: ??? https://t.co/exTyCakFxe when I invited @StringChaos to speak I wasn't even aware that he had joined @Cursor_ai fulltime; I was just hoping for the story of LiveCodeBench. Instead we got more than I ever dreamed of; a comprehensive look at all the ways in which koding evals have evolved in the last 3 years and everything that the community has learned, including o3's reward hacking, and now benchmarks to migrate REALLY large amounts of code and online evals to adjust for latency and real world constraints. ends with some actually prescriptive recommendations if you're also building coding evals of your own.

achieve ambition with intentionality, intensity, & integrity i made: - @dxtipshq - @cognition - @sveltesociety - @aidotengineer - @latentspacepod + @smol_ai

avatar for swyx
swyx
Tue Dec 16 01:04:02
  • Previous
  • 1
  • More pages
  • 666
  • 667
  • 668
  • More pages
  • 5634
  • Next