Dee Tewari

GitHub 联合创始人 Scott Chacon 的新项目 GitButler 刚刚宣布完成 1700 万美元 A 轮融资！ GitButler 是一款为现代 AI 编程工作流打造的创新型 Git 客户端（支持桌面端、CLI 以及全新的终端 TUI）,专为 AI Agent 时代设计 — 不只是给人用，也给 AI 用。 它不是"更好的 Git"，而是在重新思考下一代软件应该怎么被构建。 它的核心优势在于打破了传统的分支工作流，支持并行分支（Parallel Branches）和堆叠分支（Stacked Branches）。开发者可以在同一个工作区同时推进多个功能的开发（例如边修 Bug 边做新功能），彻底告别繁琐的 stash 和分支切换。此外，它深度集成了 Claude 等 AI 助手，能自动生成分支名、提交信息，并支持无限撤销与轻松的提交编辑。 在 UI 层面，GitButler 的设计极具现代感与实用性：采用独特的水平滚动视图，将未分配的更改、多个虚拟分支并排清晰展示，视觉层级分明。

Native macOS desktop app for local AI on Apple Silicon github.com/jjang-ai/mlxst…

I was trying to understand what #TurboQuant is all about and I found this article very helpful. pub.towardsai.net/googles-turboq… Given the recent release of Bonsai 8B 1bit quantised model from @PrismML I'm trying to understand if using TurboQuant KV optimization together with Bonsai would help push the context size further. Still new to all these but what exciting times!

I’ll show you how you can run these on Mac 16GB RAM and less too! Stay tuned

Another blow to Anthropic! Devs built a free and better Claude alternative that: - runs locally - works with any LLM - beats it on deep research - has Cowork-like capabilities - connects to 40+ data sources - self-hosts via Docker, and more. 100% open-source (20k+ stars).

Today, we demoed @PrismML's Bonsai 8B model in @AnythingLLM desktop to see if the 1-bit mode arch is the revolution local AI needs. Turns out, this is not an April fools joke. 1.1GB memory footprint for a whole 8B model with high accuracy

@greghintner @PrismML I run models using @LocallyAIApp . Great app. Does a great job of managing ram and bringing frontier models from @huggingface . Its available on the app store. Enjoy 😎

@ezeedeetee @PrismML What app do you use here to run individual models local on iPhone?

Today, we are emerging from stealth and launching PrismML, an AI lab with Caltech origins that is centered on building the most concentrated form of intelligence. At PrismML, we believe that the next major leaps in AI will be driven by order-of-magnitude improvements in intelligence density, not just sheer parameter count. Our first proof point is the 1-bit Bonsai 8B, a 1-bit weight model that fits into 1.15 GBs of memory and delivers over 10x the intelligence density of its full-precision counterparts. It is 14x smaller, 8x faster, and 5x more energy efficient on edge hardware while remaining competitive with other models in its parameter-class. We are open-sourcing the model under Apache 2.0 license, along with Bonsai 4B and 1.7B models. When advanced models become small, fast, and efficient enough to run locally, the design space for AI changes immediately. We believe in a future of on-device agents, real-time robotics, offline intelligence and entirely new products that were previously impossible. We are excited to share our vision with you and keep working in the future to push the frontier of intelligence to the edge.

Spin Up Linux Virtual Machines Quickly with KVM Using Cloud Images When most people start with virtualization, they reach for tools like VirtualBox or VMware. You install the software, open the interface, create a new virtual machine, attach an ISO, and go through the operating system installation step by step. It works. Linux also provides a native virtualization stack built around KVM and QEMU. With it, you can work directly with virtual machine disk images and build systems in a faster and more flexible way. Instead of performing a full operating system installation each time, you can start from an existing image, prepare it for your environment, and run it as a virtual machine. This approach is common in real-world environments where systems need to be created quickly, tested, replaced, or rebuilt without repeating the same installation process each time. In this guide, you will use KVM to build a simple lab environment by importing and working with disk images. You will go through the process step by step, from preparing the image to creating the virtual machine and accessing it. Continue reading: blog.sysxplore.com/p/spin-up-linu…

COBOL on a MacBook → powered by Java. Using the FFM API, you can connect legacy systems without full rewrites. This is a simple but powerful modernization path. 👉 buff.ly/hymYlLR #Java #OpenJDK #COBOL #Dev

Try our 1-bit Bonsai 8B on @LocallyAIApp

@PrismML Running on airplane mode with zero connectivity, the speed and precision of replies is really impressive. Just eyeballing the inference speed on iPhone 15 pro, I’d say it does feel like 3x - 5x faster than qwen 3.5 4B model. Mind blowing 🔥🙇‍♂️❤️

@PrismML Going to try it in iPhone now and Mac tomorrow. Very excited about this. GPUs with upto 16gb vram can now be very very productive.

Try now the ultra-efficient 1-bit 8B parameters Bonsai model, the first model from @PrismML. This model is optimized to be lightweight, fast, and energy-efficient, running on a wide range of iPhones — a first for an 8B model. Powered by MLX. Available on iPhone and iPad.

@meaning_matters @swiftcraftuk Played a bit with llama.cpp on iOS too! The best way to get started with MLX on iOS is to play directly with the swift examples repo (github.com/ml-explore/mlx…). And yeah MLX can run llama, you can try my app @LocallyAIApp if you want to quickly run it on iOS!

If you have a Thunderbolt or USB4 eGPU and a Mac, today is the day you've been waiting for! Apple finally approved our driver for both AMD and NVIDIA. It's so easy to install now a Qwen could do it, then it can run that Qwen...

Ollama is now updated to run the fastest on Apple silicon, powered by MLX, Apple's machine learning framework. This change unlocks much faster performance to accelerate demanding work on macOS: - Personal assistants like OpenClaw - Coding agents like Claude Code, OpenCode, or Codex

Git Worktrees are the secret behind multi agent systems Because they make true parallel execution possible Yet most people know nothing about them... In this 22 min video, i explain Git Worktrees and why Agents love them:

Sub-millisecond VM sandboxes are here. Zeroboot boots preloaded environments, snapshots them, then forks new isolated VMs in ~0.8ms. This changes how we think about running agents and serverless workloads.

Chrome just became massively more agent-friendly 🔥 Your real, signed-in browser can now be natively accessible to any coding agent. No extensions. No headless browser. No screenshots. No separate logins. Just one toggle to enable it. Check this out: developer.chrome.com/blog/chrome-de…