Mario Zechner

New blog post, wherein I beat a dead horse for the last time. mariozechner.at/posts/2025-11-…

Smart mode in Amp now lets you use Opus 4.6 wth 300k input tokens of context, up from 168k. Also, tokens above 200k are now charged at the same (not higher) per-token rate. Why not 1M tokens in smart? Quality/cost aren't right. If truly needed, use (hidden) large mode.

People of pi. Do you feel experimental? Want to try a new edit tool? Stuff this into your ~/.pi/agent/extensions folder. Use it with your preferred model(s) for a while. Report back if it works. Example: GPT 5.4 prefers rewriting entire files sometimes over doing multiple small edits. This mostly solves this for me. gist.github.com/badlogic/30c35…

OSS weekend starting Friday morning CET. Then I can clean out the issue and PR trackers over the weekend.

People of pi. Cloudflare release. Also, breaking changes for extension authors! Easy to migrate, just point your clanker at the CHANGELOG.md. I rewrote the keybinding mess into something that's mostly sane. Now you (your clanker) also have to suffer a little.

@rockorager @bleuonbase @mitchellh i'll take it with your token in it too, no problemo!

@bleuonbase @mitchellh I will have to clean it up to make it public since im pretty sure my token is hardcoded 😂

I'm dying for pi-mono-style minimal library that handles the hard parts of email (auth, syncing with local state, etc.) and gives me an opinionated way to add agentic loops on top of that. I want to build my own agents and logic and guardrails, I don't trust vendors right now.

@threepointone yes yes yes!

@badlogicgames Me me me

who here will be a AI Engineer London in April? I'm ready to have more pub visits.

@SIGKITTEN @mattlam_ @tobi @0xSero

@badlogicgames @mattlam_ @tobi @0xSero do your thing gramps ill wait

Keep seeing people, like @tobi @0xSero , rave about pi-mono, and with the rebirth of GUIs, thinking to make a Pi desktop app. Is there already one that I don't know about? @badlogicgames looking to make a pi-core-server similar to codex-app-server, please lmk if complete crap idea :) will be clanker slop

@mattlam_ @tobi @0xSero websocket server, allows you to: - list/delete/create/resume/join sessions (multiple clients can mutate the same session concurrently with queuing) - command interface similar to the current rpc mode - other stuff i haven't thought through yet

@badlogicgames @tobi @0xSero Makes sense, can you share any high-level constraints you already expect for integrating the future pi server, even if the design isn’t settled yet? For example: - preferred integration boundary: package APIs vs subprocess/JSON mode - whether sessions/threads will be first-class

omg it is happening and its the finest bespoke slop.

@zeeg @0xblacklight longer context is basically a hack in all current implementations. don't see this being fixed anytime soon. so longer context with the current hack won't help at all.

@0xblacklight Yeah I agree it’s both recall and quantity and neither problem improved much. Codex skill calling is fairly impressive on recall but it still struggles more as time goes on just like everything else

1) not surprising whatsoever 2) this is exactly what I keep saying about models not being powerful enough today the fact that they can do so much with lossy compression is amazing, but there's no magic here imo (for transformers) context windows need to be 1-2 orders of magnitude larger for the future people keep saying is reality, and even then the compute is probably not worth it

We've been experimenting with pi-autoresearch to optimize the new canvas rendering engine. 10x improvement on the slowest test in couple hours. 🤯 While it did take some shortcuts that degraded rendering quality, it also came up with several great ideas that were easy to cherry pick. This is huge. HT to @karpathy for the original inspiration and @davebcn87 for the π extension.

Small Pi subagents update: new option to run subagents from a fork of your current place in the convo (not just a fresh blank context), so they can inherit the context you already built up. pi install npm:pi-subagents github.com/nicobailon/pi-…

@qtnx_ @animegfz feels bad, man :( 🫂

@badlogicgames how it feels when i see @animegfz using claude

just found out my SO is cheating on me

@AuerSusan @paxaral @sudbalaji what susanne said

@paxaral @badlogicgames @sudbalaji I _think_ the issue is it messes with cache. Similar to why inserting an updating time stamp in the system prompt is a not advisable.

this is immensely illegal

@jdkornac @VictorTaelin i don't need it. i just want people to leave me alone. didn't work out so great and here we are ...

@badlogicgames @VictorTaelin That’s very surprising, you came up with a great extension system but you never use it? If I may ask, why?

Ok so I thought that was a dumb gimmick but now I'm completely sold on how pi is a self-modifiable software. It literally knows how to modify itself very cleanly and that's extremely useful in practice I'm not using Codex / Claude Code anymore Bend2 should definitely be like this! I mean, constructed in a way that AI's can easily navigate it and know how to modify it to add any feature the user wants. Perhaps we're past the era of open source software and into the era of forkable software, where the most hackable project wins?

Here's the pi-coding-agent UI you weren't asking for 😘📷 @badlogicgames Mucho interactivo. Colored dots = agent sessions. Shared canvas for your daily dose of 🐐♥️🫠 #AI #DevTools #pi

@tharshan_09 i can't bring myself to tell her about cowork, or i'll lose her forever.

@badlogicgames haahahah. How do they like claude cowork? or they prefer the cli?