Learn more about custom agents and what they can do for you. 🤖 https://t.co/7FXtMTVCLb

How do we make software more secure for all of us? Log4Shell proved that open source security isn’t guaranteed and isn’t just a code problem. It’s about supporting, enabling, and empowering the people behind the projects that build our digital infrastructure. Watch the full interview with Log4J maintainer Christian Grobmeier: https://t.co/oFxr5efh2L

If you believe in home devices that respect privacy, give you freedom and still play nice together, then you should probably read this. https://t.co/X45kRn35MH

FUCK https://t.co/mVldmLCT3z

Organizations running large-scale inference now have a clear path away from single-vendor lock-in—without giving up performance. MAX on AMD MI355 delivered: - 2× throughput - 40-60% cost reduction - More tokens/$ across real workloads Published benchmarks with @tensorwave: https://t.co/nWBeQJ3wUj

Revealing the Coder Color of the Year… https://t.co/vevOKC6uth

Ever wondered how the agent system prompt works in @code? Me too. So I dug into it over the weekend and made this just for you. https://t.co/wYDgVNDMMy

I think most Europeans are actually completely ignorant about their track record for arrests for social media posts, probably because their press is captive and the data is almost always heavily censored / not released. But here are the facts: - UK: >10k arrests per year for "offensive communications", basically mean tweets. in 2022 alone, 77k "public order offences" investigated by police. by far the worst offender. - Germany: hides and obscures the data. but 4,486 cases of Volksverhetzung (incitement to hatred) in 2018. most of these are online posts. this means we're talking a few thousand arrests for online speech per year - France: hides the data. about 3000 cases a year for racist offenses that are purely speech-related and carry no actual physical harm (“injures, diffamations, provocation à la haine”) In the US, posts for merely offensive speech are not criminal. Yes, you can still be arrested for making specific threats, stalking, etc. But you won't get arrested for calling a politician fat, as happens in Europe. These are plain facts.

Software Engineer (Research and Development) https://t.co/E1uPJa5tIy

Business Role (Corporate Development) ビジネス職（経営企画部） https://t.co/2VbQQ7FCzw

Made this video to explain evals https://t.co/8VmviZLHmR

over 50,000 models with API providers on Hugging Face https://t.co/pAMcAoSqI4

https://t.co/xgZrJVXSE0

Zhipu AI just released GLM-4.6V on Hugging Face This new multimodal model achieves SOTA visual understanding, features native function calling for agents, and handles 128k context for documents. Perception to action! https://t.co/SbX2zr5raT

Meta AI's Saber redefines zero-shot reference-to-video generation It generates stunning, identity-preserving videos from text & images. No costly R2V datasets required, trained solely on video-text pairs. Achieves state-of-the-art with masked training. https://t.co/KCO1ZwddpE

docs: https://t.co/FQXswxLpQ5

We just launched @CTGTInc's Mentat, an OpenAI-compatible API that gives enterprises deterministic control over LLM behavior. Benchmarks showed clear gains in accuracy, truthfulness, and hallucination prevention. We built it after seeing models ignore correct information or produce misconceptions because of the internal patterns they rely on during generation. Prompts and RAG could not correct that. Mentat uses our policy engine to govern behavior at the feature level in real time. It aligns output with an organization’s rules and trusted information and resolves inaccuracies before they reach the end user. This reduces hallucinations and keeps answers consistent without fine tuning or fragile prompt stacks. Full breakdown here: https://t.co/DmmsB2ATNK

What's the matter babe you've barely touched your Chinese translation of hillbilly elegy https://t.co/RgnZncrncs

What's the matter babe you've barely touched your Chinese translation of hillbilly elegy https://t.co/RgnZncrncs

having close male friends is so crazy https://t.co/VBlo5nRRUg

having close male friends is so crazy https://t.co/VBlo5nRRUg

https://t.co/P2ucDWq054

+2 more

WATCH: Small plane crashes into car while landing on I-95 in Brevard County, Florida https://t.co/WpAFd2INs4

Starlink Direct to Cell is now delivering data, voice, video and messaging in Canada. Customers traveling in the most remote areas can stay connected on several apps, including checking the weather, routes on maps or even posting on X using the satellite-to-mobile service → https://t.co/QCcLwOmHuE

Jason Cammisa on trying @Tesla FSD V14 during an hour long trip after being critical of FSD in the past: "I had to eat my shit and my words; This f**king car drove us the entire way, expertly, making lane changes whenever it wanted, with nothing dangerous. No jerks, no side-to-side. Beautifully smooth. It f**king pulled into the parking lot and backed itself into a parking spot. And then it drove us all the way back to San Francisco; This is door to door amazing. Couldn't believe it."

10/10 growth hackers https://t.co/u2ecSjitnx

The results are in. Just 69.5% (n=115) people at this neurips knew what AGI stands for. This is only slightly up from last year. See you again next year! https://t.co/k8lcnmEvPf

Split documents into distinct sections automatically with our new LlamaSplit API 📄✂️ We're excited to introduce LlamaSplit (now in beta), which uses AI to automatically separate bundled documents into clear, targeted sections based on categories you define - no more manual splitting of document stacks. 📋 Analyze page content and classify pages into your defined categories with natural language descriptions 🎯 Get back precise segments with exact page ranges and confidence scores for each section ⚡ Handle real-world scenarios like resume stacks, mixed financial documents, court filings, and research paper collections 🔗 Combine with LlamaExtract to run targeted extraction on each segment or route to appropriate agent workflows Perfect for processing resume bundles,handling mixed document types, legal teams organizing court filings, categorizing patient charts and more. Watch an example of segmenting (an AI generated) bundle of resumes below 👇 Read the full announcement and get started with LlamaSplit: https://t.co/cQqeZCGeww Docs: https://t.co/mvZtbRVIBn

I did not expect that the PowerPoint killer would be something called Nano Banana Pro, but that is where its heading It makes the major efforts by all the other AI companies, including Microsoft, to crack PowerPoint by using python seem like a dead end ImageGen is all you need? https://t.co/xMe68cuEYp

you can just prompt with your hand: https://t.co/DRq0OrFyHo

you can just prompt with your hand: https://t.co/DRq0OrFyHo

Browser agents used to be for engineers only. Not anymore. Astro from @asteroid_inc is an AI agent builder for the hardest legacy portals, letting anyone automate them in minutes. Proven in healthcare, insurance, and finance. Congrats on the launch, @MlcochDavid & @simulated_land! https://t.co/Y8aOfwYy0w