π ToolsΓ1 mentions
Quinn Image 20
βYou can try it at chat.quinn.ai.β
youtube: ByteDance's Seed Dance 2.0 Redefines AI Video Realism Amid Surging Multimodal and High-Speed LLM Advances β by Matt Wolfe β 2026-04-12
Endorsed by Matt Wolfe
What the smart people are recommending. 8874 books, tools, and products endorsed by the thinkers absorb.md tracks. Ranked by how many times each has been recommended across compiled podcasts, papers, posts, and tweets.
βYou can try it at chat.quinn.ai.β
βThe easiest way to use it is to go to gemini.google.com. google.com. Click the little create an image button...β
βyou can try it over at app.quiver.ai and they give you 20 credits.β
βUm, from a consumer standpoint, I use ChadByt heavily.β
βI would say is very similar to let's say how the Linux community approaches things which is you need to build new features you need to rapidly fix things you need to make changes like you need to stayβ¦β
βDeepSeek v3.2 just came out and I saw that they have sparse attention as an example, and this is one way to have very, very long context windows.β
βRunway Chooses Modal to Power Real-Time Inference for Runway Charactersβ
βvery excited about the upgrade GaussGym provides for environments for training locomotion capabilitiesβ
βThe quantum virtual machine can be deployed in one minute from a colab notebook where you can program your quantum circuit colab notebooks are super user-friendly and packed with great features for quβ¦β
βSo things like Gemini, VO, imagine uh etc.β
βI hope you do pip install uh circ and if you do so you'll get the latest and greatest version and we look forward always to getting feedback from you about cirqueβ
βCode and models will be released at https://xiaobenli00.github.io/ETCH-X/.β
βIt's been exciting to use your model inside Slack with our new Slackbot.β
βWe also updated tensorflow quantum with some new features adding for example matrix product state circuit simulationβ
βSo if you haven't tried, uh, give the new Gemini Flash latest and Gemini Flashlight latest models, uh, a shot. They're pretty good.β
βSo I'm a big fan of Comfy UI and nodebased interfaces in general >> and that is complex and it's complex but it's also it's very robust and you can do a lot of things and so you know after we releasedβ¦β
βDeepagents deploy helps you do that, easilyβ
βUm I'm sure you've had it's it's been insane. And it's so funny like I now half of my X feed is these really heavy Nano Banana users in Japan who have created like Chrome extensions called there's oneβ¦β
βThe terminal-to-html tool should include a checkbox for if the created Gist should be private, and it should default to checkedβ
βLangSmith's new Insights Agentβ
βMulti-turn Evals in LangSmithβ
β# How to prompt Veo 3.1β
βProject Genie: Experimenting with infinite, interactive worldsβ
βThe three options right now for you and it's different in Europe because I know there's limitations but that's Claude 3 Opus that gp4 which you can also get access to for free in a limited form througβ¦β
βthe president signing the u stablecoin bill into law which was the the genius act. Um and I I think that I'll just tell you what we see is like the positive consequences of that law have been even bigβ¦β
βLabelbox can get you rubrics tailored to your domain, helping you systematically measure and shape how your models think. Learn more at labelbox.com/dwarkesh.β
βThe dataset was built using a novel framework, DeepCluster++, designed to ensure intra-class diversity and reduce manual curation.β
βThe model is available now on @GoogleAIStudio and the Gemini API.β
βI would buy Groβ
βThe three options right now for you and it's different in Europe because I know there's limitations but that's Claude 3 Opus that gp4 which you can also get access to for free in a limited form througβ¦β
βAnd we can put those together in a hybrid simulation engine that we open source called Robocassa. And it is an engine that leverages LOM's diffusion models text to 3D to procedurally generate infiniteβ¦β
β- [numpy](https://numpy.org/install/) <3β
β- `datasets` for huggingface datasets <3 (if you want to download + preprocess OpenWebText)β
β- `tiktoken` for OpenAI's fast BPE code <3β
β- `wandb` for optional logging <3β
β- `tqdm` for progress bars <3β
βI recommend getting the bleeding edge PyTorch nightly ([select it here](https://pytorch.org/get-started/locally/) when installing) as it is currently quite likely to make your code more efficient.β
βThis downloads and tokenizes the [OpenWebText](https://huggingface.co/datasets/openwebtext) dataset.β
βOne thing is maybe kind of random, but like I get really fired up to see like mad science experiments like the uh Deepseek OCR that came out the other day. Did you Did you see it? It's It's wild whereβ¦β
βI saw another thing on hacker news the other day where um you know uh text diffusion \nuh where someone made a text diffusion model by instead of doing go saying dnoising he would take like a single Bβ¦β
βMistral Large v2 is now compatible with `mistral-finetune`!β
βthe open zone map that we can put on screen. The open zone map shows hundreds and hundreds hundreds of special economic zones globally, right?β
βThe code and model are open-source and accessible via Spinal Cord Toolbox v7.0.β
βWe also introduce a lifelong learning framework to automatically monitor the morphometric drift as the model is updated using additional datasets.β
βHere we present RareCollab, an agentic diagnostic framework that pairs a stable quantitative Diagnostic Engine with Large Language Model (LLM)-based specialist modules that produce high-resolution, inβ¦β
βClaude Opus 4.7 is now the default orchestration model powering Computer.β
βUsing glm-5 as a daily driver for a lot of thingsβ
βI first got to know his work through his blog which can be found at one useful thing. orgβ
βhe was getting a lot of value out of the chatbt desktop integration with his terminal and that it was a very simple thing. He was just like, if there's an error in my terminal, I just ask chatbt like β¦β
βbut like the u the gro 2.5 um open source model is actually very good. Um, and I think we'd probably be and and we'll continue to open source our modelsβ
βIn fact, we did the math and published it published the math, but nobody looked at it. Uh um it's on the Tesla websiteβ
βThat's Gbrain, the Y Combinator CEO's personal AI knowledge brain, now fully open-sourced for everyone to use and build on. GitHub link in the description.β
βthe best tool out there in tutoring is Khan Academy's kigo um which while flawed is still the best approach out there right now to doing tutoring with AI and like anyone can subscribe for 20 bucks a mβ¦β
βthen Quen is the model from Alibabaβ
βToday we're releasing Personal Computer.β
βNamex is a simple utility to separate the implementation of your Python package and its public API.β
βJust use deepagents and the memory built in there!β
βI created an organization called IP which which is Iran prosperity project but it first focus is the first 100 days. How we immediately stabilize the situation in Iran.β
βIf you want your agent to research anything, I'm talking products, competitors, news, market data, literally anything, plug this in now.β
βYeah, so Lyria is our music model, um, and we have some really interesting experiences to, to bring the music model to life. Um, so if you go to inside of AI Studio, if you go to the Build Tap, there'β¦β
βThe last project I did, I actually was like, "Oh, I can actually just install Cloud Code into a GitHub code space." And all of a sudden, I have Cloud Code against the repo that I've already been buildβ¦β
βAt one point, they even worked with Mosaic ML, another past guest, to train their own code model.β
βsimple but solid cursor and VS Studio support for when you need to call in a professional human developerβ
βBut one of the cool ones was around, we released a, a Gemini 2.5 flash text to speech model. Um, and actually you can do a bunch of really. Interesting product. Uh, um, I, I can send a link for folks β¦β
βI can prove in pen and paper algorithms that we wanna run that like, it's not that we don't know what to do with a quantum computer. There are hundreds of algorithms you can go to. I think it's calledβ¦β
βYou can create a template in your framework like you know nest next.js JS, Web Flow, WordPress doesn't, you know, just stick with the thing that you're going to use.β
βHref for for a while has had this backlink check, uh, checker.β
βYou can use tools like Otterly, there's profound, there's manual testing.β
β# Run Isaac 0.1 on Replicateβ
βIn this work, we introduce Ξ-Motif, a GPU-accelerated subgraph isomorphism algorithm that reformulates the task through the lens of database operations.β
βThis is the one nobody talks about, but everyone should be using. Automated security testing for your AI prompts, red teaming, edge case testing, vulnerability scanning.β
βI as soon as I get home this weekend, I'm going to be in Firebase Studio because I love just the deep integration with Firebase and just how seamless everything works.β
βI found 03 to be a step change in practical usefulness where like typing a question into a box and then seeing what it says or dumping a bunch of information into the box and asking what it thinks or β¦β
βInstall this before it happens again.β
βSkills MP, over 500,000 skills, largest catalog by far.β
βSkill hub, 30,000 plus skills that are AI-rated, so you can actually sort by quality instead of just guessing.β
βthis is a really cool tool called map of GitHub. Um this is kind of like all the software that's written.β
βIf you want to stay on the bleeding edge without spending your morning scrolling GitHub trending, bookmark this.β
βFirst up, Yi Combinator President Gary Tan has open- source GStack, an AI coding workflow system for cla code.β
βwhen you just change the URL from GitHub to get ingest then uh this will actually concatenate all the files into a single giant text and it will create a directory structure etc.β
βMaybe some of you know three blue one brown. He makes beautiful animation videos on YouTube... he wrote uh Manon and I wanted to make my own and uh there's extensive documentations on how to use manonβ
βscottarensson.comblog. That's my blog.β
βIn this case, we start with the assumption that our charting library (`recharts`) doesn't support server-rendering.β
βYou can use tools like Otterly, there's profound, there's manual testing.β
βBy the way, Graedia is awesome. Okay, great. I mean, just phenomenal.β
βwe put out GPT3 in the API and the world didn't care, but sort of Silicon Valley did. They're like, 'Oh this is kind of cool. This is pointing at something.'β
βback in 2019 before the rise of LLMs I released an AI benchmark to highlight this difference. it's called the abstraction reasoning corpus or ark1. And from that time back in 2019 to now...it really rβ¦β
βthat tool is ARGI 2 which released in March this year...ARK 2 challenges reasoning systems...The benchmark format is still the same. There's a much greater focus on probing compositional generalizatioβ¦β
βwe're going to launch this in early 2026 and next month in July we're going to release a developer preview so you can start playing with it.β
βcrons, part of langsmith deployments! https://docs.langchain.com/langsmith/cron-jobsβ
βThese findings position NQS-SC as the new default approach over NQS-VMC for electronic structure calculations.β
βwe have been like obsessed with claude co-work, but mainly OpenClaw. And so what we did was... we had 15 of them come in this weekend and they all got trained over like six or seven hours had to have β¦β
βWe have Kimmy on them. We are moving all of the work onto these and then they'll use Kimmy for most of their easy jobs, which is free.β
βOne of my partners at Craft actually created a new tool called Lobster Tank, which is a version of Open Claw that's got some enterprise security wrapped around it.β
βOne was Jules and one was called Stitch if you came to the developer keynote. And those are kind of our hypotheses really of what it might look like if you had a coding agent as a suite, a professionaβ¦β
βbenchmarks leaderboards like LMSIS right and there are a couple more as well like web dev arena or like the scale leaderboards and things like that those are important for us because they are still caβ¦β
βAndre and Horowitz has a really I think a really good uh bar chart where they compare the companies of the uh early and mid to late 80s to uh today where I mean it just basically demonstrates through β¦β
βI even took a test for childhood trauma and scored alarmingly high. You can do this yourself. It's called the adverse childhood experiences test or ACE for short.β
βSo I invite you to go to uh the website. Uh this is the QR code. You can it's free. You can just download it and play with it.β
βI invite you to give command and control a try if you're interested.β