Skullthoguhts

Sci-Hub is an evil website that pirated 85M+ research papers and made them freely available And now they've added AI to their database to make Sci-Bot. It answers your questions using latest, full-text articles. But DO NOT use it. We should all try to make billion-dollar academic publishers richer. I'm putting the link below so you know how to avoid it.

@d1nt @NainsiDwiv50980 thank you .

@commandlinetips @NainsiDwiv50980 youtube.com/watch?v=E1ytrE…

This 2-hour Stanford lecture will teach you how to build AI agents, write prompts, and implement RAG — better than every Claude and AI tutorial you've watched this year combined. This is exactly how Stanford trains its own engineers. Free. On YouTube. Bookmark this before you forget. 👇

@parcadei i will thanks

I've not used that, so I can't vouch for whether it'll 100% work fine but I would just add it and test, since that look to only work on compressing command output which is different to editing there is the FastRead tool, which outputs structural information about a file and already pre-compresses when reading because it uses tldr, and a fast diff tool but I don't know if that part is going to clash with sqz compressing it's output It might not, but I would just add and test it - if they do, let me know if there's issues and I'll see if I can help

you're still paying full token for edits? meet FastEdit the tool to slash your edit token spend, and free you from usage limit misery fine tuned merge model with a tldr engine free, open source and swaggier than your corpo overlords parcadei.github.io/fastedit/

Karpathy didn't make a course. He made THE course. 3 hours. Free. Tokenization. Attention. Hallucinations. Tool use. RLHF. DeepSeek. AlphaGo. Every behavior you've ever wondered about in an LLM - where it comes from, why it exists, how it was engineered. The gap between engineers who understand this and engineers who don't isn't technical depth. It's the ability to conceive of entirely different things.

@Lenovo when will see this on legion ?

The workflow isn't incidental to the quality it's structural. the process is the product.

the power of open source

TheRealHumans: THE MOVIE

@lexrus what about the sounds of typewriter how did you do it thanks

NeoVIM + Herdr + gitu + Ghostty(with shaders) + Keeby * Please turn up the volume.

@brendonovich great news i hope your eyes will be fine after the rework it will be perfect for it to have the wterm future renders to the DOMand and sqz github.com/vercel-labs/wt… github.com/ojuschugh1/sqz

we've moved opencode desktop to electron. it's faster, more reliable, and will replace our tauri build soon. try it out in beta via the link below.

You're Bleeding Tokens With Your MCP Tools Here's How I Fixed It open.substack.com/pub/commandlin…

@TareqAmin_ How we can test the Human AI I would like to test it and compare to other models is there a way for an API for usage

Huge appreciation to the HUMAIN Compute team for the relentless commitment shown over the past few days. Operating around the clock working on the assembly and activation of the AI compute B300 stack is no small task. I am especially proud that this has been achieved fully in-house. No external dependencies. No consultants. Just our team, our expertise, and our execution. This is what @HUMAIN stands for: ownership, speed, and excellence at scale. #HUMAIN #TheEndOfLimits

@dhh just have time to watch the Omacon and loved the MicroSlop logo

Omacon is streaming live x.com/i/broadcasts/1…

The eye running that if you are free user you get free quote everyday for some messages but if you are a paid user with monthly subscription you cannot use it if you hit your limit funny .

@Adidotdev respect <3

Why is he still pushing code ?

@thdxr dax pleas the GLM5.1 have input text and image the open code server don't accept images its a config thing in the opencode server don't show to the model the image in the chat pleas fix it

what if we gave you unlimited tokens for free and we also paid you

Canceling my Claude subscription. this week was ugly hitting the limit constantly and i barely use it. light user and already unusable. the models are better, for real. but what's the point if i can't use them. this is my last month .

@vikassabbi sure DM !

@commandlinetips i can send you code if you wanna try

After the #Claude #Code source code #leak, most of us are going to build our own #CLI tools , While we are at it there's a way in there that'll save tokens It's called Code Mode (Cloudflare's approach) , Instead of 15+ MCP tool definitions eating your context window,