Brendan Toscano

CS grad and AI researcher

[Brendan Toscano @AIEngineer10x]

@jun_song Bet 8gb of ram and 5tb of vram lol

[Brendan Toscano @AIEngineer10x]

@sama I think usage limits

[Brendan Toscano @AIEngineer10x]

@trq212 @theo They are definitely watching your videos.

[Brendan Toscano @AIEngineer10x]

@melvynx Not much cause it will expire soon

[Johnny Jnr @jturntdev]

Glad everyone else is noticing that Codex usage limits have been destroyed.

>It’s draining as fast as 5hr limit. But weekly
>Cheaper models, are seeming to drain our quota just as fast.
>20x plans feel like 5x

“5.6 is extremely efficient, and costs just as much as 5.5” yeahh right.

[Brendan Toscano @AIEngineer10x]

@jturntdev @thsottiaux Feels like more users come at the cost of our usage and the resets are basically the same as anthropics fake promises.

[Johnny Jnr @jturntdev]

Sorry but Codex Usage limits are terrible now?

Yesterday, i only used 5.6 sol high, i let it running in a /goal, configured cheaper terra/ lunar subagents.

I come back 40% of my total weekly usage is gone? ( 20x)

BTW i was using multiple 5.6 Sol xhigh + Max, all day everyday, last week.

Id rather get no resets, steady and fair usage, as opposed to getting resets, but every time they do, they reduce our quota.

They also removed the invite friends for banked resets.

Anyone else noticing this? Just gets worse by the day.

[Brendan Toscano @AIEngineer10x]

@thsottiaux I need the "thing" to be able to use it.

[Brendan Toscano @AIEngineer10x]

@LexnLin @openai Just Do it.

[Brendan Toscano @AIEngineer10x]

@miniii_codes @grok context pls

[Brendan Toscano @AIEngineer10x]

@thsottiaux We are all waiting, pls wake up.

[Brendan Toscano @AIEngineer10x]

@droidbuilds I was praying I don't see spotify. Literally anything but spotify hell even Amazon music.

[Brendan Toscano @AIEngineer10x]

@TheAppleDesign Its a symbiotic relation, to think one is better than the other shows you don't understand how business works. Samsung is the main display (other than LG) supplier for the upcoming iphone. Samsung was the one of the first to pioneer the folding display tech. @grok

[Brendan Toscano @AIEngineer10x]

@a_fresh_shinobi Double tap the power button. Single tap to silent

[Brendan Toscano @AIEngineer10x]

@thsottiaux @rudrank What's the point of banked resets if we only get normal ones.

[Brendan Toscano @AIEngineer10x]

@RayanKrishnan I couldn't care less if they hacked anthropic and stole Fable 6 and all its training data. F anthropic W kimi

[Brendan Toscano @AIEngineer10x]

@Kavan2908 Coincidence?

[Brendan Toscano @AIEngineer10x]

@niyonx Try Netcup, don't use contabo no matter what people tell you.

[Jun Song @jun_song]

New prediction :

We will be able to run Mythos level local AI in 4 months.

On a single Macbook with 32GB RAM.

Mark this and come back.

[Jun Song @jun_song]

This is a prime example of the info gap in local AI.

Most people think Qwen3.6 27B is the best, so they run it on a 128GB Mac.

But MLX is actually optimized for MoE, not dense models, fully leveraging that massive unified RAM.

Running a 27B dense model is obviously going to be painfully slow and frustrating.

Meanwhile, running the dq version of Minimax-M3.0 or Deepseek-v4-Flash is a way smoother experience.

The catch? No frontier AI can tell you this. You can't search for it, and knowledge cutoffs mean they don't have the latest facts.

This exact info gap is why people keep saying local AI is years behind.

Just look at the latest info on experts' X accounts and follow what they do.