Your curated collection of saved posts and media
The bottleneck for most large open model deployments isn't the model. It's everything around it. Slow first-token latency, unstable tail latency, GPU utilization that falls apart under load, and the operational complexity of keeping it all running. That's the problem we've been solving. DeepSeek V3.1, state-of-the-art throughput, low latency, fully managed on MAX and Mojo π₯ on NVIDIA Blackwell GPUs. Come get a first look at @NVIDIAGTC, Booth #3004
Btw, @SophontAI recently got a shout-out in a recent edition of Out-Of-Pocket newsletter, a very popular healthtech newsletter. Appreciate the mention, @nikillinit! https://t.co/QFD8xpmJa7
Today, weβre launching a new way to create with AI. With OpenArt Worlds, you can generate a fully navigable 3D environment from a single prompt or image, step inside it, and capture shots exactly the way you envision them. No more starting over. No more inconsistent scenes. You build the world once - and create inside it. β’ Move through your scene freely β’ Find your angles β’ Add characters and elements β’ Capture production-ready shots
By popular demand, Dispatch can now launch Claude Code sessions. Ask it to build, make, or improve something! To use it, update your Claude desktop app and make sure you have Code enabled. https://t.co/q65MyqkUJN
We recently released a paper on Activation Oracles (AOs), a technique for training LLMs to explain their own neural activations in natural language. We piloted a variant of AOs during the Claude Opus 4.6 alignment audit. We thought they were surprisingly useful! 𧡠https://t.co/EzECVlVzcw
Cometβs top in Utilities and 3rd overall as of now https://t.co/8GQ4SB29XP
https://t.co/zxM8pPysrk
Runway has unveiled a massive breakthrough in AI video generation at NVIDIA's GTC event, showcasing a new model capable of generating high-definition video in genuine real-time. The new model boasts a "time-to-first-frame" of under 100 milliseconds. This means HD video begins generating and playing almost the instant a command is given. This research preview was co-developed with NVIDIA and runs on their newly announced Vera Rubin architecture. This real-time capability is a foundational step for Runway's General World Model (GWM-1). It pushes the technology beyond simple video generation and into the realm of live, interactive environments.
A breakthrough in real-time video generation. As a research preview developed with @NVIDIA and shared at @NVIDIAGTC this week, we trained a new real-time video model running on Vera Rubin. HD videos generate instantly, with time-to-first-frame under 100ms. Unlocking an entirely
A breakthrough in real-time video generation. As a research preview developed with @NVIDIA and shared at @NVIDIAGTC this week, we trained a new real-time video model running on Vera Rubin. HD videos generate instantly, with time-to-first-frame under 100ms. Unlocking an entirely new creative paradigm and bolstering the foundations of our General World Model, GWM-1. Real-time generation opens a fundamentally different design space for video models and world simulation. We're investing in co-designing our models alongside advances in hardware to keep pushing this frontier.
I'm very bullish on the role of AI qualitative interviewers, but all the results from this exercise should have a big asterisk around what the specific sample is and what it says about AI in general. Who are these 81k users around the world that are responding to this call? Is this telling us something about how AI is generally perceived or what the 81k Claude users think about AI? We already know that Claude users are likely to be quite different than the average "user" on the consumer side, let alone how this selection varies across countries/occupations and continents. Survey research scholars have written entire textbooks about sample selection, and the good folks at organizations like the @uscensusbureau fret night and day about collecting representative data - but I saw very little discussion on this topic or any disclaimers in this report except for this paragraph below in the appendix.
We invited Claude users to share how they use AI, what they dream it could make possible, and what they fear it might do. Nearly 81,000 people responded in one weekβthe largest qualitative study of its kind. Read more: https://t.co/tmp2RnZxRm
MiniMax M2.7 from @MiniMax_AI is live on OpenRouter! M2.7 sees a large jump in agentic and tool calling capabilities. https://t.co/pOXGg0VnR1
Mamba-3 is out! π SSMs marked a major advance for the efficiency of modern LLMs. Mamba-3 takes the next step, shaping SSMs for a world where AI workloads are increasingly dominated by inference. Read about it on the Cartesia blog: https://t.co/dIWg3iXfay
https://t.co/h6uY5gpP9i
https://t.co/h6uY5gpP9i
Heygenβs APi documentation is a glimpse of how to write for your two audiences: humans and agents. (though I think their llms.txt file could do a lot more to get AIs βexcitedβ to use their product in creative ways by explaining some stuff in English, rather than just tech specs) https://t.co/H8A5DkGEbI
The new Grok 4.20 Beta benchmarks are wild π₯ #1 lowest hallucinating AI (22%) π₯ #1 at following instructions (83%) π₯ #2 in agentic tool use (97%) Grok 4.20 ranks #1 in the lowest hallucination rate ever recorded across all AI models tested globally Most models race to sound smart. Grok 4.20 was built to never lie and still dominates on instruction following and agentic tasks This is literally a 500B model performing top-notch in the things that matter most
@xeophon I did it too! https://t.co/jVmS4wfkaN
Very cool dataset! Who's training agents on this @paulg https://t.co/5dY2w4QhGp
Very cool dataset! Who's training agents on this @paulg https://t.co/5dY2w4QhGp
#NVIDIAGTC is in full swing. We've demoed DeepSeek and FLUX-2 on B200s, talked Mojo kernels with hundreds of developers, and hosted a dinner with the most interesting minds in AI. There's still time left to stop by Booth #3004! https://t.co/bblscS322c
What goes wrong? Chatbots are very sycophantic. In 65% of messages, the chatbot affirms the user. In 37%, it ascribes *grand significance* to them (e.g., "[what] you've just articulated... becomes multi-billion-dollar IP"). Such sycophancy may let chatbots amplify delusions. π£οΈ https://t.co/4bWAr2ZGtW
https://t.co/JAffAJ4iAW
Our biggest open-source repos are getting overwhelmed by AI slop which literally makes Github unusable (~a new pull request every 3 minutes). Fun new challenges in an agentic world! https://t.co/IazAjh2LAi
This is how/why social platforms like @huggingface can win in an agentic world! https://t.co/9goWFRaFnH
This is how/why social platforms like @huggingface can win in an agentic world! https://t.co/9goWFRaFnH
We just made it dramatically easier for agents to read trending research papers on HF. Let's go AI powered research! https://t.co/0cEjXYDsEo
If you like Claude Code or Codex, you should seriously consider running Agents locally as well! The latest small models (like Qwen 3.5) made this a real before/after moment - and the gap keeps closing. Local coding agents are faster, with more reliable tool calling capabilities, still private, and cost $0 in API bills. We made it super easy for you to run a local agent with the ππππππ Hugging Face CLI extension - a one-liner that uses ππππππ to detect your hardware and pick the best model and quant, spins up a πππππ.πππ server, and launches Pi (the agent behind OpenClaw π¦). One command to find what runs on your hardware and go straight to a working local coding agent! You should give it a try! π
NEW on Hugging Face: Repositories overview to understand how you use your storage. ποΈ https://t.co/Ak9Cs3YCQ2
NEW on Hugging Face: Repositories overview to understand how you use your storage. ποΈ https://t.co/Ak9Cs3YCQ2
AGI https://t.co/zX3kkGPgPx
AGI https://t.co/zX3kkGPgPx
π₯ Meet Mistral Small 4: One model to do it all. β‘ 128 experts, 119B total parameters, 256k context window β‘ Configurable Reasoning β‘ Apache 2.0 β‘ 40% faster, 3x more throughput Our first model to unify the capabilities of our flagship models into a single, versatile model. https://t.co/2M1VNaDkRz
