Your curated collection of saved posts and media
Wow, such a great open-source drop! Put Claude on steroids with 120+ scientific skills spanning maths, biology, chemistry, medicine, engineering, & more 🤯 If you’ve ever wanted Claude to act like a research assistant, this gets very close. Free and open source. Repo in 🧵↓ https://t.co/gWRt3Kzr37
totally blown away by Google Gemini 3 + three.js.. just generated a 3D product interactive app with zero coding.. use hand gesture to rotate, scale, control xyz and lighting of 3D model AI is taking over.. accept or not tutorial with prompts & code below: https://t.co/l3R99B1GEu
Introducing @imagine_dev_ The AI builder with Appwrite Cloud built-in. Build something real. https://t.co/MRzNMdzCcq
Introducing @imagine_dev_ The AI builder with Appwrite Cloud built-in. Build something real. https://t.co/MRzNMdzCcq
Introducing GLM-Image: A new milestone in open-source image generation. GLM-Image uses a hybrid auto-regressive plus diffusion architecture, combining strong global semantic understanding with high fidelity visual detail. It matches mainstream diffusion models in overall quality while excelling at text rendering and knowledge intensive generation. Tech Blog: https://t.co/tVpAVsSmxb Experience it right now: https://t.co/M9UaDtxGoV GitHub: https://t.co/u7HpXQ1WCJ API: https://t.co/SOOzRdErVS Discord: https://t.co/hFcRgVZAqJ Coming soon to https://t.co/DRCwFHQj1P Chat and GLM Coding Plan.
Announcing our latest open medical AI models for developers: MedGemma 1.5, which is small enough to run offline & improves performance on 3D imaging (CT & MRI), & MedASR, a speech-to-text model for medical dictation. Both available on Hugging Face + Vertex AI. https://t.co/jPm4UodeAH #MedGemma #HealthAI #GenerativeAI
Glimpse of a world of fully generative interfaces. AI → JSON → UI: https://t.co/BKcvtDky5K https://t.co/QH6ctR1ldA
Glimpse of a world of fully generative interfaces. AI → JSON → UI: https://t.co/BKcvtDky5K https://t.co/QH6ctR1ldA
Today we're open sourcing a technical preview of the GitHub Copilot CLI SDK. Build agents with custom tools in Go, Python, TypeScript, and C#. Built on the same agent loop that powers the Copilot CLI and GitHub Coding Agent. Supports BYOK, and any model. Here is the Copilot CLI driving Excel:
There was a flippening in the last few months: you can run your own LLM inference with rates and performance that match or beat LLM inference APIs. We wrote up the techniques to do so in a new guide, along with code samples. https://t.co/vpNGPZ0vgM https://t.co/wCaWIXwBf6
You can now do reinforcement learning training with 7× longer context and no accuracy loss, via our new batching algorithms. Long reasoning chains in RL are costly, but now we enable you to train gpt-oss with GRPO & reach 380K context on a 192GB GPU. https://t.co/io8OUqGIbn
we're open-sourcing specstory cli we’ve spent a lot of time normalizing agent sessions, and defining a stable contract for ai conversations across tools. and now we are open-sourcing it would love to see what people contribute, build on top of that and build better systems around provenance, memory, and tooling. https://t.co/h8eExqxtf9
🚀 Ling Studio is officially live — the official web platform to interact with Ant Ling models, offering 500,000 free tokens every day to explore, test, and build. In Ling Studio, you’ll experience Ling-1T’s lightning-fast responses, explore Ring-1T for powerful and stable reasoning, and try Ming-flash-omni-Preview to explore image and audio understanding - all in one place. Key features: * Flexible Ling / Ring configuration (parameters + system prompts) * Native tool calling, including web search * API-ready for easy integration into your applications Coming soon: file-based chat, image generation, a growing set of model skills, and more.
AntAngelMed is now open source. 100B parameters. The largest open-source medical LLM to date. Built on the Ling-flash-2.0-base MoE architecture. Ranked #1 among open-source models on OpenAI’s HealthBench. https://t.co/EzNDKdIvQU
even your toaster can see now 🔥 @ultralytics dropped YOLO26 family of models, not only for detection.. 30 models for open vocab detection, segmentation, keypoint detection and more! 🥹 all models <50M params, this demo runs on CPU 🤯 https://t.co/WT8R7zE71V
We’re releasing TranslateGemma, a new family of open translation models with support for 55 languages. 🌐 Available in 4B, 12B, and 27B parameter sizes – they’re designed for efficiency without sacrificing quality. https://t.co/SRJzCOAKyG
Maps codebase architecture from a GitHub URL https://t.co/9Xm6HXgVUW https://t.co/p68HTEdBF5
Maps codebase architecture from a GitHub URL https://t.co/9Xm6HXgVUW https://t.co/p68HTEdBF5
R.I.P. basic prompting. MIT just dropped a technique that makes ChatGPT reason like a team of experts instead of one overconfident intern. It’s called “Recursive Meta-Cognition” and it outperforms standard prompts by 110%. Here’s the prompt (and why this changes everything) 👇 https://t.co/caDTB52TS9
tinygrad 0.12.0 is out! No more ShapeTracker, AMD wave visualizer with VIZ=2, NVIDIA support without CUDA, and MI300/MI350 driver. https://t.co/df84HVaaGa
🐦bird 0.7.0: fast X CLI for reading tweets: now with home timeline, news/trending, user-tweets and plenty fixes. Thanks @albfresco @odysseus0z @gakonst and all other contributors who didn't list their Twitter. 😊 https://t.co/O4HZcOYt10
Vibe Coding Robotics Part 6 Built a Theo Jansen's Strandbeest simulator to see how an AI models handle complex linkage systems Built with Gemini 3 UI generated with Nano Banana More details ↓ https://t.co/khuXGY9go6
Here's why most attempts to resize windows on macOS Tahoe fail: https://t.co/vDsJtQYjxo https://t.co/gxZAwmmQda
Library of engineering skills for AI agents https://t.co/Su5dTRj9Ec https://t.co/XUFvRgdIuO
Library of engineering skills for AI agents https://t.co/Su5dTRj9Ec https://t.co/XUFvRgdIuO
🚨 NEW: Apple granted new patent that allows for a technology that eliminates the need for prescription lenses in upcoming AR/XR glasses by correcting all prescriptions including astigmatism on device. https://t.co/gw4YgYIMAm
Cautious Weight Decay is a surprisingly simple technique that has been repeatedly validated in Modded NanoGPT. I expect it will gain serious traction as the default variant of decoupled weight decay https://t.co/BLFSeqrf2w
modded-nanogpt WRs 🤝 Cautious Weight Decay Pretty much all the @speedrun WRs since November have used CWD in some form (e.g., @varunneal's "CWD w/ schedule"). Huge thanks to everyone experimenting and sharing results — shoutout to @varunneal, @classiclarryd, @ChrisJMcCormick,
Did some statistics. My productivity ~doubled with moving from Claude Code to codex. Took me a bit to figure out at first but then 💥 https://t.co/cfyKg0E1hf
Someome made a morning report skill and I just gave @clawdbot the tweet and it set up the skill + cron job. https://t.co/CXo0xMGcFv
Cowork but with local models not to send all your data to a remote cloud! https://t.co/2OrBMMO3NJ
In Cowork, you give Claude access to a folder on your computer. Claude can then read, edit, or create files in that folder. Try it to create a spreadsheet from a pile of screenshots, or produce a first draft from scattered notes. https://t.co/GEaMgDksUp
Cowork but with local models not to send all your data to a remote cloud! https://t.co/2OrBMMO3NJ
Codex team is at it again with just another insanely useful feature. If you see your agent going off the rails, or needing some addt'l context, you no longer need to stop the agent. Follow up prompts while the agent is working now inserts the prompt at the next thinking step, allowing it to pivot or provide better outputs without actually stopping the agent. This is undoubtedly a token and time saver. You can still queue up prompts using Tab. Well played!
