Havriil Pietukhin

@benln Bratislava when?

Cafe Cursor in Kraków

@badlogicgames fixing vibed python code takes a lot of tokens

Jebus, what are ya all doing with all those tokens? at this rate, we'll be top 3 in two weeks.

@badlogicgames 540 mb idle vs 160 in pi is still crazy

IT IS HAPPENING!

@aliouftw leet

i love pi.dev 's tree and fork but i also like having a clean context when moving from investigation to implementation. so i made a `/spawn` command that simply creates a new child session with the last message of the parent session bonus, looks nice in `/resume`

someone wrote a 680 page interactive book on cs algorithms

@alexeeps @alexisbchz again, how did you compare some js runtime and programming language?

@pantasw2 @alexisbchz java is 10x more bloated than bun will ever be

bun is on the verge of becoming the most bloated runtime ever (don't get me wrong, i love bun)

@alexeeps @alexisbchz what does it have to do with java?

@alexisbchz ever heard of java?

🤗🤗🤗introducing Hugging Science -- the home of AI for science 🤗🤗🤗 open models and datasets are the powerhouse of science (see the PDB), but finding the models and data you actually need for your breakthrough is hard af you shouldn't need to scrape arxiv, own your own wetlab, fight a custom HDF5 parser, build a fusion stellarator, and beg for compute before you've trained a single epoch so we're changing that we've put all the best science on @huggingface in one place: - 78GB of genomics data - 11TB of PDE simulations - 100M cell profiles - 9T DNA base pairs - 13M molecular trajectories - 400k medical QA pairs and much more, all open, and all ready for training (+ you can also now filter and search by domain, task, and keyword) we've put together all the biggest releases from our partners at NASA, Google, OpenAI, Meta FAIR, Arc Institute, Ginkgo, SandboxAQ, Proxima Fusion, NVIDIA, Ai2, OpenADMET, InstaDeep, Future House, Polymathic AI, LeMaterial, Earth Species Project, Merck, and Eve Bio if you're not sure where you fit in -- work on open challenges for problems that matter: including fusion stellarator design, ADMET, antibody developability, multilingual medicine, catalysis and materials, and scientific reasoning. we're already changing how science gets done: a fusion startup needed a benchmark for stellarator plasma confinement that didn't exist. @proximafusion shipped ConStellaration on Hugging Science: a leaderboard, dataset, and eval metrics, all in one place. a drug discovery team wanted to predict hPXR induction. OpenADMET put up a blind challenge: 11,000+ compounds assayed at Octant, 513 held out, two tracks (pEC50 + structure). Anyone in the world can train and submit. an antibody team at @Ginkgo released GDPa1, a developability dataset for stability, manufacturability, and immunogenicity prediction, with a live leaderboard scoring every submission. if you know a problem the ML community should be working on, let us know. make a challenge! this is about putting all the tools for solving science in one place. so we can hillclimb! → huggingscience.co

@badlogicgames long live the king Mario!

compare tokens to requests. what does that tell you?

@badlogicgames @wesleimade it also runs perfectly on Raspberry Pi Zero

@wesleimade actually running on the phone? or ssh session to somewhere?

How is this even possible, @badlogicgames? Two instances of mono-pi running on my grandfather’s old phone (Samsung Galaxy A03)! Amazing work!

🇭🇷Croatia is receiving the largest investment in the history of Central and Eastern Europe, the Pantheon Project worth 50 billion euros is underway. An AI development and innovation center, known under the working name Project Pantheon, will be built in the area of ​​Topusko in Sisak-Moslavina County, with a total installed capacity of one gigawatt. The value of the infrastructure alone that will be built through this project and will remain with the state - from new power lines and substations to road and optical networks - is estimated at around 500 million euros, more than the Pelješac Bridge cost.

@michalkvasnicak plannotator.ai

is there any diff tool that allows me to add comments and runs the claude code / codex / opencode / pi once i apply the comments?

@badlogicgames @mitsuhiko at first i wanted to criticize font choice, but now it beautifully clicked for me. now i have only 1 regard - small text in docs navigation is too pixelated imho

People of pi.dev. We have a new website. It's a still a bit WIP, but our humongous web team is working on it day and night. You can now also update pi itself via pi update. Thank @mitsuhiko for the perfect font choice.

@dmwlff already used this cc version, pretty neat pi.dev

After 2.1.117, you may notice that Claude doesn't call its Grep or Glob Tool anymore. YES!!! It only took four months. It's faster than ever and it's all Bash. It's so much harder to take things away than to add them. Enjoy.

@badlogicgames @lucasmeijer great! looking forward to it. where do you stream?

@pantasw2 @lucasmeijer no need to donate. maybe i stream sometime later this week when i do issue triage.

People of pi. Cloudflare release with a minor breaking change. - JSON validation no longer relies on AJV, but uses Typebox. Update your existing extensions as per changelog (@sinclair/typebox -> @typebox) - OSC 9;4 indicator in supporting terminals for @lucasmeijer thanks to MVP @cinnamon_msft - Tools can now signal "terminate the loop" so you can do structured output and other shenanigans easily - Custom auto completion providers via extension. This one fetches the last 100 isues from GH when i type # and let's me search through them via type ahead And lots of bug fixes and internal refactoring.

@badlogicgames total MLX win

clampy clampy clampdown. just waiting for OAI to clamp down as well.

@Vtrivedy10 great framing, thanks i hope that pi.dev will eat the world

there’s clearly some confusing conflicts + double-think going on in AI between: 1. Closed labs saying use our harness, it’s naturally post-trained and gets the best out of models by being “in-distribution” 2. “Models are approaching AGI” and generalizing to do incredible work across many domains 3. In-context learning maximalists pushing skills repos and special instruction sets 4. Task specific harnesses and models vastly outperforming “general purpose models” and harnesses if I had to pick between any of these I say - you can build a task specific harnesses and outperform the closed lab harnesses, evidenced by some work we’ve done, Factory, terminal bench, etc - models are not incredible generalization machines, they’re very sensitive to their post-training distribution but that’s prob fine, we can build harnesses around them to do useful work - the future is owning your model + harness stack and deploying specialized intelligence so that you don’t need to rely on a big lab for everything. big labs will still produce great models not all of the above 4 points can be true and every player sort of has an interest in a subset of those being true among consumers, model labs, Agent labs, open source model makers, vertical ai companies. it’s totally worth doing evals for yourself here more open models + harnesses means you at least have a choice which path you choose 🚀

Always choose optimism over fear