Introducing T3 Code - Roles.
I cut a release of my fork which introduces the concept of "roles" for subagent delegation.
Assign specific models to task categories to avoid constantly repeating "Create an Opus 5.5 xhigh subagent for design review."
Live on Linux and Mac
So, I went down this path far enough to create a t3 fork with a dedicated "lead/orchestrator."
One version with persistent threads and another with ephemeral threads.
I ultimately killed it.
T3 in it's current form handles the task fine. Coordinators ultimately are just there to reconcile and spot drift when needed, which is a single ask that you can invoke as a thread when needed, then settle.
Once you get used to the rapid thread creation and settlement of t3 you learn how to conform your approach and I've yet to spot degradation.
The one thing that my fork experiments did reveal for me though is that I want more default controls over the models chosen for subagents and alt-threads.
I hate having to say "this is design, use opus 5.5" 67x a day, or that subtasks should use Deepseek.
So the one forked feature I'm still iterating on is a default model based on "roles" where I can opt in to a preference for subagents for parent level threads and the parent knows who to delegate to based on the task at hand.
Early days but results are promising. Doubly so because I inject ambitious pre prompts to the subagents automatically.
Replying to @adi_baradwaj @aiamblichus
Why can't you tell them to unblock themselves or coordinate amongst one another?
1
1
90
Replying to @0xJiro
Same, although it feels like innovation pace is bringing it within the range of hardware I already have
46
Replying to @RealAZKicks
Not wrong about most. Still, maybe we could have >0 even from the well intentioned group
99
Replying to @Zodomo
I'm less technical then you so it's nice for me. Herdr and Orca I hear are better and those who enjoy TUI based dev
67
Replying to @pratikg @TheDavidTai
Was thinking about doing it from my m4 pro w 24gb but need to see if it's worthwhile
49
Replying to @Zodomo
All via T3 which utilizes their native underlying CLIs.
So Claude Code, Antigravity and Codex
1
2
136
It's funny. All the gaming NFT projects from 2021 died because it cost wayyyyy too much in $, expertise and human hours to ship a good game.
You give a motivated group tens of millions in the AI age and they can probably ship an incredible game.
Looks like wizards community is trying to revive theirs and no doubt it's because AI lets them do more with less on top of all the assets the original mints paid for.
No more waiting; we launch November 30th.
With tons of changes and adds for old and new players, the Forgotten Runiverse Game is back and reborn.
Check back for updates leading up to the date.
We look forward to seeing you in the server, Wizards!
7
33
3,816
Eff. I might actually have to start local model maxing.
We worked with @thinkymachines to make Inkling 2 Mini run fast locally:
- 210 tok/sec decode, 3620 tok/sec prefill on 96GB M5 Ultra Mac Studio
- 125 tok/sec decode, 1750 tok/sec prefill on 48GB M5 Max MacBook Pro
It built a native weather app 1.8x faster than Claude Sonnet 5.5 on an M5 Ultra Mac Studio and 11x cheaper (API pricing).
The looping architecture is ideal for local devices like MacBooks, which are memory constrained. Looping layers of the model increases inference-time compute (and intelligence) without increasing the size of the model.
3
26
2,720
Replying to @ChimeraDefi
Got it:
And a breakdown by token + $ spend.
Google models archieved more with less, despite using more tokens.
Of the 3 groups, only Google's lead model had less token spend than its deligates. Astra and Fable spent more delegating than their underlying spent testing + prototyping.
1
60
And a breakdown by token + $ spend.
Google models archieved more with less, despite using more tokens.
Of the 3 groups, only Google's lead model had less token spend than its deligates. Astra and Fable spent more delegating than their underlying spent testing + prototyping.
Gemini won.
I gave Astra, Carbon and Fable the same prompt for finding performance gains on the a codebase, with the only restrictions being:
→ Feature parity
→ No UX degradation
→ 12hr time limit
Astra basically didn't do anything, and stopped after 2hr. It's "big idea" brought key presses from 1.3ms to 0.5ms (lol).
It was really efficient though (193 loc).
Fable innovated on some responsiveness metrics, missed on the other metrics, and then spit out 3x the amount of code for test scripts. Definitely the "safest" to merge if I were to one shot it. It also reward hacked somewhat by showing a responsiveness metric hitting 250ms while the CPU finished on the backend over 2s so the "gains" were still felt on the system.
Gemini was the only model to take the guidance prompt to heart, which specifically said large structural changes are okay if nothing is lost. It achieved strong gains across the board, but also didn't test as thoroughly as Fable. This was apparent when it caught a few edge cases issues like 15min before submitting.
It was also just very hard to track throughout its process. The text is much more code-dense than Fable or Astra and can be hard to follow (pictures at the end for comparison).
In the age of agentic coding, I want ambitious models so the win here goes to Gemini. I'll be giving it the job of reviewing all submissions and coming up with a merge plan to preserve as many gains from all 3 as possible.
So far, really pleased with Google's new model.
Replying to @OanaGoge
I wanted to pit "flagship" against one another for this.
This is a good call out though and I've assigned Neon the super merge management
2
3
333











