我们的 LLM 会讲话了，time-to-first-audio (TTFA) 小于 300ms blog.lepton.ai/voice-mode-com…

不错呀，小模型干大事

Bolt.new 宣布更新💥 现在可以在任意 Github 仓库 前面加上 bolt.new 即可在 Bolt.new 中打开该仓库🔥 示例：bolt.new/~/github-vaeyed 打开之后即可选中代码做提问，并且能够立即预览！后续还能支持基于代码进行问答搜索和帮助你在线编程🤔

New 3h31m video on YouTube: "Deep Dive into LLMs like ChatGPT" This is a general audience deep dive into the Large Language Model (LLM) AI technology that powers ChatGPT and related products. It is covers the full training stack of how the models are developed, along with mental

azure.microsoft.com/en-us/blog/emp…

This is how 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗠𝗲𝗺𝗼𝗿𝘆 works. In general, the memory for an agent is something that we provide via context in the prompt passed to LLM that helps the agent to better plan and react given past interactions or data not immediately available. It is useful to

太酷了 ⚡️ 清晰可视化的解释 LLMs 的函数调用和 MCP 🔥 带非常直观的动效视觉效果 💥 （建议收藏

合适的工具组合

中国速度

还记得你何时加入 X 吗？我知道！#我的X周年纪念日

We are the Harry Potter generation!

Top 10 YouTube channels to learn AI from scratch: 1) Andrej Karpathy - a blend of general and technical content, Zero to Hero playlist is a must-watch youtube.com/andrejkarpathy Andrej Karpathy 2) Umar Jamil - highly technical, implements ML and LLM techniques from scratch

受用

这次发布的还不少，慢慢研究一下 文心大模型技术Blog （含技术报告下载）： yiyan.baidu.com/blog/posts/ern…文心一言： yiyan.baidu.comHugging Face： huggingface.co/baiduGitHub： github.com/PaddlePaddle/E…百度智能云：

现在 AI 的发展已经到达了一个非常尴尬的时刻。 各种模型发布，提升微小，官方运营、kol、用户都快测不出区别了 各种创始人访谈，访谈都吹的天花乱坠，产品一发布完全不是一回事 各种教程，可能是最有用的内容了，但是太多了，根本学不过来。 圈内人，已经快跟不上了。 圈外人，就直接跳过了。

Done with copy-paste translation? Here’s a step-by-step guide to building an automated pipeline with just a few lines of Python: 🔎 PaddleOCR → parse docs with high accuracy 🌐 ERNIE 4.5 → deliver nuanced translation A real productivity boost for multilingual workflows. Full

这回我同意他的观点😎😎😎

ERNIE X1.1 just launched! Unveiled at Wave Summit 2025, our latest reasoning model reduces hallucinations, improves instruction following, and delivers strong agentic capabilities. ✅ Achieves significant performance gains over ERNIE X1, with factual accuracy up 34.8%,

近期一直在看这方面的资料，想动手做，但是一直感觉agent在RAG这方面好像没那么明显的作用，新思路打开新界面

有点惊艳看完