MoreRSS

site iconFrost Ming修改

Python 开发工程师,pdm 作者,坐标深圳。平时喜欢折腾技术,写写技术文章。也是个开源爱好者。
请复制 RSS 到你的阅读器,或快速订阅到 :

Inoreader Feedly Follow Feedbin Local Reader

Frost Ming的 twitter 的 RSS 预览

我做了个梦,把credentials.json没ignore掉提交到PR里了,过了半天才发现

2025-05-30 11:46:47

我做了个梦,把credentials.json没ignore掉提交到PR里了,过了半天才发现



yetone: 这两天的阿里云开源项目中硬编码密码事件再次验证了我之前的这个看法:就不应该在设计配置文件的时候把硬编码密钥内容这种东西给设计进去!

https://mp.weixin.qq.com/s/ucbH1I2rcTVpsB4YuVA4KQ

RT Visual Studio Code: Today, we're announcing plans to make VS Code an open source AI editor. We believe AI development should stay true to VS Code's...

2025-05-20 00:09:40

RT Visual Studio Code
Today, we're announcing plans to make VS Code an open source AI editor.

We believe AI development should stay true to VS Code's core principles: open, collaborative, and community-driven. Let's build the future of software development together.

https://aka.ms/open-source-ai-editor

??

2025-05-11 09:01:10

??

RT 面包🍞: 我的《重学前端》课程正式发布了,包含了最新的TypeScript、Raect19、Next.js 15……预期每周一集。 同学们可以点击下方的链接看第一集 https://ww...

2025-05-05 06:32:18

RT 面包🍞
我的《重学前端》课程正式发布了,包含了最新的TypeScript、Raect19、Next.js 15……预期每周一集。
同学们可以点击下方的链接看第一集

https://www.youtube.com/watch?v=dQw4w9WgXcQ

以为改个名字就不用还了吗,我不管,它就叫司母戊

2025-05-04 14:43:15

以为改个名字就不用还了吗,我不管,它就叫司母戊

RT BentoML - Infrastructure for Building AI Systems: #BentoFriday 🍱 — Model Composition Many real-world problems are too complex for a single mode...

2025-04-26 00:17:00

RT BentoML - Infrastructure for Building AI Systems
#BentoFriday 🍱 — Model Composition

Many real-world problems are too complex for a single model to solve.

Whether you’re building #RAG pipelines or #AI agents, you’ll often need to combine multiple models, each doing part of the job. It also helps you build more accurate, flexible, and robust AI systems.

But orchestrating and scaling these components, managing hardware allocation, and ensuring performance? That’s where things get tricky.

Why developers love #BentoML for building multi-model pipelines:
✅ 𝗥𝘂𝗻 & 𝘀𝗰𝗮𝗹𝗲 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗺𝗼𝗱𝗲𝗹𝘀 𝗼𝗻 𝘀𝗲𝗽𝗮𝗿𝗮𝘁𝗲 𝗵𝗮𝗿𝗱𝘄𝗮𝗿𝗲 (GPU & CPU)
✅ Process different types of data with different models
✅ Orchestrate models 𝗶𝗻 𝘀𝗲𝗾𝘂𝗲𝗻𝗰𝗲, 𝗶𝗻 𝗽𝗮𝗿𝗮𝗹𝗹𝗲𝗹, 𝗼𝗿 𝗯𝗼𝘁𝗵
✅ Improve performance with ensemble techniques
✅ Ideal for 𝗺𝘂𝗹𝘁𝗶-𝗺𝗼𝗱𝗲𝗹 𝗮𝗴𝗲𝗻𝘁𝘀, 𝗥𝗔𝗚, 𝗮𝗻𝗱 𝗺𝘂𝗹𝘁𝗶-𝗺𝗼𝗱𝗮𝗹 𝗮𝗽𝗽𝘀

🔗 Learn more: https://docs.bentoml.com/en/latest/get-started/model-composition.html

Here is a simple #InferenceGraph example 👇