That was fast https://youtu.be/TVOoMwkpSRQ

0 reply

0 recast

3 reactions

When agents change to this behaviour as default, MCP will really explode. https://youtu.be/Xs2CkHEpIrM

0 reply

0 recast

1 reaction

We're in hell. https://x.com/SawyerMerritt/status/2010878861202931814 @asenderling.eth @tbsocialist @chainleft

0 reply

0 recast

1 reaction

OpenCode Zen/Big Pickle is very very solid Been using in Zed while I’m on Claude timeout and it’s been doing really well so far.

1 reply

0 recast

2 reactions

Can someone explain context windows to me? Because I read that I could just switch to working on a different project or querying Claude about everyday stuff (thus switching to a new context), but when I try, it enforces the rate limit no matter what I want to do.

2 replies

0 recast

8 reactions

I’m imagining a scenario where you give AI access to your encrypted messaging app, and it decides that it would easier to parse messages if it decrypted them and added them all to a networked attached database in plain text form.

0 reply

0 recast

4 reactions

Are there any open models that work in a really similar way to Claude Sonnet 4/4.5? I haven't found any other LLM whose agentic coding approach is as well-designed. ChatGPT 5/5.1/5.2 creates so many headaches every time I use it. 4.1 is ok, but Claude really takes the cake imo. But yeah, open models. One of my goals this year is to switch to running my own models, and I'd love something as similar to Claude as possible.

3 replies

0 recast

6 reactions

If you don’t regularly tell your LLM that it’s wrong, you are being led astray significantly. The reality of this, which represents that these are simple computing tools which still require task-specific knowledge and expertise, is what worries me most about stupid (or at least overconfident, uncritical) people using LLMs.

4 replies

2 recasts

27 reactions

The models in Raycast output so much faster than the same models in their own chat interfaces. Anyone else noticed that?

2 replies

1 recast

17 reactions

Good video. Twists and turns, covers a lot of ground. Mostly, it’s an indictment of corn. 🌽 https://youtu.be/H_c6MWk7PQc

3 replies

0 recast

32 reactions

PSA: You can use Nano Banana Pro in Venice using your credits.

0 reply

0 recast

2 reactions

I appreciate that they’re trying to make ChatGPT less sycophantic, but now it’s telling me bluntly that it can’t do something it’s just done a few minutes ago…

3 replies

4 recasts

25 reactions

Someone took all the currently released Epstein files and loaded into an LLM so you can ask it to parse anything from the files openepstein.com

1 reply

1 recast

2 reactions

Ok I need to know. I always use ChatGPT (or Copilot) for web design experiments—and some paid work. It can accomplish most everything I want it to, because I know enough about HTML, CSS, and JS to know what to tell it to tweak. But I feel like I’m fighting it. It reverts to prior mistakes, it gets slower and slower and I lose a lot of time and become very frustrated. What should I be using instead? Or is 5.1 still better than the competition?

4 replies

1 recast

7 reactions

hfs. stay safe out there https://x.com/brave/status/1980667351000584627

1 reply

1 recast

8 reactions