parth.

just look at this beauty man.. i want ts

@dabit3 @DevinAI quick question, as i was using handoff i noticed the model got set to opus 4.7? is there a way to change it to smth like kimi or anything else?

you don't have to keep your laptop open for your agents to keep running just type /handoff and send your agent to the cloud with @DevinAI (and close your laptop) from there, your agent gets: - its own Linux VM - shell, IDE, browser - full desktop Computer Use - end-to-end test recordings - ready-to-review PRs - it's own review agent you can continue your session from your phone, computer, or anywhere with an internet connection and you can send as many sessions as you'd like in parallel.

@0xSero what skills do you mainly use sero?

Do you want to learn to use AI, and contributed your session data to open source so we can train better models? Models better than Opus We need as many people as possible to contribute their agent traces from their claude code + codex history Pi's Mario & I both shared ours.

@MattSchrage @cognition need to be verified for that 😭

@TotallyNotParth @cognition /bug or just DM me

@cognition's Devin CLI came out. Written in Rust. Compared to other CLIs, which are mostly Node.js / TypeScript slop. Rust wins on performance, memory safety, and zero runtime overhead. Cognition definitely made the right call. Subagents can also be spawned. Infinitely.

@MattSchrage @cognition Found a issue! Any way i can contact you? Shouldn't be too major but it shouldn't be happening either

@TotallyNotParth @cognition thanks for trying it out! let me know if you have any feedback

@NousResearch Used tools: Hermes Agent by @NousResearch Devin CLI by @cognition Codex by.. (cmon, self explantory) Models Used: GPT 5.5 in Codex & Hermes Agent Kimi K2.6 in Devin CLI.

Nurse Joyless: Nurse Joy heals Pokémon. Nurse Joyless heals bad decisions. > A competitive Pokémon Showdown team clinic for structure analysis, matchup diagnosis, KO risk, replay evidence, and hidden-information reasoning. @NousResearch

@NousResearch Very rushed Demo Vid, so apologies for that!

@Teknium sincerest condolences, glad you're focusing on yourself tho!

Ok, home from the traveling for the funeral. Got some rest. Feeling at peace and ready to rip it

@tibor_tee @aye_aye_kaplan handsome boi

Thanks to @aye_aye_kaplan's fun automation sporting a new profile pic in Slack

@arshi_mohib seems like that's going to the bookmarks! I'll be watching. guarantee you keep pushing on and you'll make it! & The Linux / Windows integration.

@TotallyNotParth Will gift you a new one my friend if Screenforge reaches $5K MRR 😌

Greatest mistake of my life was buying a 256GB Mac.

@AravSrinivas perfect is probably an understatement, that guy looks pretty damn good

The perfect cube doesn’t exi—

@NebulaAI could you check my dm? I've reported some bugs.

Introducing Public Agents It’s the easiest way to share an agent you’ve built, or install one someone else made Teams are already using agents to: - audit and secure apps - run interviews/screen candidates - run operations for businesses - and 100+ examples Try it now for $0

@0xSero framework laptop as in THE frame dot work !?

First REAP on a Framework Desktop/AI Ryzen Max Compression on Framework is cheap. Image below: Saliency scoring and heat-maps, on the right you can how often experts are selected, the lighter the hotter. A lot of the model is not very relevant for a handful of domains. CHOP

@m0n3ch been saying

chrome is actually a virus

@0xSero hook me up too dell

Dell, hook up your boy

@ryanvogel @KevinNaughtonJr @FrameworkPuter you should hook him up with a 13 pro (and me just a regular 13)

@KevinNaughtonJr this should be me

never kys

I still absolutely love using @NebulaAI , it's one of the first tools i used after getting back into AI. Currently using v4 pro with it & i sometimes forget how good it is when it's just testing things for you. all the best to the nebby team!