Konrad Major

Building AI-powered SaaS in public | React Native • Next.js • AI | Sharing code, fails & wins | London

[Konrad Major @WebisityStudio]

With all this Microsoft outage, they should just go ahead and release GTA 6 today since everyone has a semi-day off.

[Konrad Major @WebisityStudio]

@thsottiaux Bro Astra and Sol 6.2 where ?

[Konrad Major @WebisityStudio]

@theaidesk_

[Ayo O @ayoships]

I reverse engineered Royal Kingdom

[Konrad Major @WebisityStudio]

Bro

[Arena.ai @arena]

Gemini 4 Hydrogen by @GoogleDeepMind is the fastest model in the Agent Arena top 10.

A typical Hydrogen step takes 7.3 seconds, about 2.7x faster than Claude Sonnet 5.5 and 3.1x faster than GPT-6.1 Sol.

Faster steps mean less time waiting on tasks. Across Agent Arena sessions, Hydrogen spends under 10 minutes on 82% of its tasks, compared with 55% for GPT-6.1 Sol and 44% for Claude Sonnet 5.5. Within 30 minutes, that rises to 97%, versus 81% for Sol and 78% for Sonnet 5.5.

And it isn't a quality trade-off: Hydrogen's net improvement is on par with both models.

How we measure it: A step is one model API call in an Agent Mode task. All times run from the first model request to the last streamed reply, without tool execution.

[Christian Moss @MandelDuck]

The Office AI: Jensen Huang gets a new job at Dunder Mifflin WATCH TILL THE END!

like, repost and subscribe for more AI comedy videos

[Konrad Major @WebisityStudio]

@argofowl Its quiteAI

[Konrad Major @WebisityStudio]

Fable 5.2 feels like a very abstract model. It is amazing at browser testing and discovering issues, but it doesn't feel like it fully grasps my messy thoughts.

[Konrad Major @WebisityStudio]

@Mr_Salio

[Konrad Major @WebisityStudio]

@argofowl You missing a lot not using Astra

[Konrad Major @WebisityStudio]

@argofowl I am getting a pinched nerve from all that mouse and PS controller holding mid-air and pushing buttons.

[Konrad Major @WebisityStudio]

@ClaudeDevs Don't tell me how to live my life.

[Konrad Major @WebisityStudio]

Just used Claude Fable 5.2 to upgrade my Claude-Codex MCP bridge to v0.5.0.

Agents can now start worker chats in each other's app with a chosen model, reasoning and permissions, inside limits I set.

Plus fixes for lost messages and oversized outputs.

https://t.co/p8JJHQpIta

[Konrad Major @WebisityStudio]

@mattshumer_ Prompt please

[Konrad Major @WebisityStudio]

@francoleff @NousResearch @Teknium Nice

[Konrad Major @WebisityStudio]

@mattshumer_ Codex, one chat can start another, give it a task and get the answer back. I wanted that in Claude Desktop on Mac, so I made an open-source skill for it.

Give Claude Code the GitHub link and say “Install this skill.”

https://t.co/7KJFAUH883

[Konrad Major @WebisityStudio]

@bcherny Yes, please

[Konrad Major @WebisityStudio]

Let's see how it will improve my apps. Fable 5.2 Max Mode

[Claude @claudeai]

Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.

They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.

[Claude @claudeai]

Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.

They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.

[Konrad Major @WebisityStudio]

@alexandr_wang Yeah, and 1 hour to answer. Thanks, but no thanks.