yizuz

@maria_rcks Let theo bee😭

the year is 2027

Is it just me, or does GPT 5.6 Sol burn through tokens incredibly fast, regardless of the thinking mode? I wasn't even doing a big task.

@jasondeanlee I think 5.6 it's more for that area but just like comparing an apple vs samsung hahaha

Which is better for math? Fable or 5.6?

I can finally talk about GPT 5.6

@keyserfaty I think you should take a break 💀

guys

@craigweiss It's becoming a cult there, so it will be worthless Hahaha

we might need to start an anthropic employee helpline

@NousResearch @Shpigford That's why I like hermes it's so nice for local Ai's

@Shpigford Hermes Desktop

the deeper i get into using chat, CLIs and MCPs for everything, the more i find myself desperately just wanting a nice proper UI. tired of everything just being walls of text.

@rezoundous Le Chaton Fat is the way 🍟

if US and China starts gatekeeping frontier models, don't worry. we still have Le Chaton Fat

@pawelk411 @maria_rcks For real

@_yizzuz_ @maria_rcks Fable worse than drugs 💀

I’m suffering from post-Fable depression

@theo I don't think so, anthropic is plaged with researchers he just want to be in his place with his people

Do you think Karpathy joined Anthropic just so he could use Mythos for ML research without restrictions?

@ash_twtz For me it's different I'll tell chat gpt to create a document use this pictures and for some reason wants to create images 🥲 happens to often or like for unrelated prompts like yours

Bro Chatgpt you shouldn’t create images for appreciation 😭

Actually I been testing kimi of hardening the vps that's is on, company workflows to see how's it performs of how generates documents on docs like a personal assistant and kimi omg it behaves like claude not 100% but I'll say around 60% but in coding task I'll say 75 to 80% on that

@_yizzuz_ Any workflow you built with hermes? or automated something?

Cheap tokens hack: run kimi k-2.6 through Cloudflare Feels close enough to Claude for many tasks and costs basically nothing. Thank me later! :p

@suryanox7 @thsottiaux Exactly pls @thsottiaux see this

@thsottiaux Reliability first always Nobody cares about agent speed if it confidently destroys a repo at 2am 😄 Features attract users Performance impresses users Reliability keeps users Especially for coding agents, trust compounds harder than benchmarks ever will 

For those of you living inside the codex app, what should we prioritize among features, reliability or performance?

never deleting this app 😭

you know what all of these "which is better" polls are silly use codex or claude code, whatever works best for you i am grateful we live in a time with such amazing tools, and grateful there is a choice

@mil000 True, who knows?👀

@craigzLiszt That's a fact right there!

“just google it” has almost entirely been replaced in my vocabulary by “just ask ChatGPT”