Fascinating from 4chin local models thread:China plans to have 7nm DUV by 2040Bro jerks off to GLM every day and it doesn't engender any self-reflection about…
RT Piotr Mazurek (in Oxford 🇬🇧)Total canon layer cultural victory 🫡
Inkling was somewhat inevitable once Tinker took off -- integration of post-training services across @thinkymachines's entire stack -- and it's one of the best open-model business…
RT Michael ElabdA model drop by Thinky 🚨🚨Have been doing some early testing on the model for the past couple of days. Here are some of…
New open weight model drop! https://x.com/i/broadcasts/1nxeLLlOrQAJX
Fascinating PokemonVincent Chow: Anthropic's head of national security policy Tarun Chhabra accused Zhipu of distilling both Claude and OpenAI's models for GLM-5.2 at the Aspen Security…
RT Horace HeIt truly takes a village to release a model, perhaps especially an open weights model. Actually doing the entire process from scratch, from data…
Re We're proud of the model, but we have a lot more to do. It's our first public model that came out of the model factory…
RT Thariqthis was such a fun talk with Cat and Simon, hope you get to check it out if you missed itAI Engineer: 🆕This Year In…
RT Design ArenaBREAKING: Inkling by @thinkymachines is 9th overall on Agentic Web App Arena by Design Arena with an Elo of 1257It's an open-weight model in…
Excited for our first general model Inkling -- open weights, 975B, natively multimodal (text, image, audio). Available on Tinker, HuggingFace and partners.It is yours to personalize…
RT NVIDIA AICongrats @thinkymachines on the new open model 🙌Inkling was trained on NVIDIA GB300 NVL72 and the NVFP4 checkpoint is available today on @huggingface: https://huggingface.co/thinkingmachines/Inkling-NVFP4Happy…
Our first model, Inkling. Trained from scratch, weights are open, fine-tunable on Tinker today.Thinking Machines: Today, we are introducing Inkling.Inkling reasons efficiently across text, image, and…
Love this approach to open releases from Thinky -- release a practical foundation to build on. I expect this to succeed for them, much as Tinker…
I yearn to see how Dario's dream turns into nightmare, as he scurries exponentially faster and yet Chyna continues breathing down his neck, the status quo…
RT BoxThe #BoxWorks agenda is live, and @nvidia CEO Jensen Huang is taking the main stage.Jensen will join Box CEO @levie for a keynote fireside chat…
RT AI Engineer🆕This Year In Claudehttps://www.youtube.com/watch?v=uU5Gv2h8-9g@simonw chats with @_catwu and @trq212 about the state of:- @claudeai Code- Claude Fable- @anthropicai culture & product strategy - Claude…
> on the scale of largest GPT post-training runsRed the SeducerA mind made solely to tempt into sinThe closest OpenAI have come to building a DevilAndrew…
RT OpenAIIntroducing GPT-RedAn internal automated red teamer on a mission to find our models’ prompt injection vulnerabilities at scale, helping us build stronger defenses before wider…
RT Matthew BermanI know I'm late but yeah ok Grok 4.5 is actually really good. It's super fast, direct path to solutions, and quite capable.
RT TeslaBuying a Tesla is easyOrder within 2 mins on our website, then complete the entire process in the Tesla appKatie Miller: “My @Tesla buying experience…
I really like thisI don't care about "fable 5" (no idea how Fable does on such tasks) but if legit, it updates me to 55ish on…
RT Andrew McCarthyThe sheer amount of energy it takes to defeat gravity is humbling, but a powerful reminder of what humanity is capable of. My photography…
Congrats to @rimelabs on the $24M Series A 🚀Rime is building the voice layer AI products need, and we’re proud to serve their models on Together…