𝐎𝐌𝐀𝐒𝐔

Filmmaker building apps |

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@chenzeling4 From disk…

What’s the output speed? I bet it’ll take 50x the time for the model to respond from disk instead of RAM.

Novel idea, tho.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@pagan_hoetry To special bookmarks you go…

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@rezoundous But financially, considering how good it is? They’re getting 🤑

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@ImSh4yy If it’s fable level… shouldn’t we treat it as such???

Just lower the reasoning level to medium. And set it up as an orchestrator 🤷‍♂️

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@sidrelll @thedulab Now that’s a good quote

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@shannholmberg Can’t we delegate reading files to a subagent? Or do you believe it would degrade Fables reasoning?

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@ElevateTSpirit Vibe coding first thing in the morning. Still play in some regard but towards creating. A sort of a middle ground.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@MarcosHernanz Cursor? Droid?

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@DeepakNesss Eyyyyy, you got it!

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@elvissun Best part? it's only gonna get better.

Anyone still remember o3 pro pricing? 150$...

Waiting for GPT Sol

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@DeepakNesss maybe better? I'm considering to back to cursor instead of codex. Interesting stuff going on, now. Pretty exciting

https://t.co/VcxOGi3P0s

[Morgan @morganlinton]

I'm on the $100/mo Claude Max plan and I have been running Fable 5, pretty much non-stop since it was released, and can't seem to hit any usage limits.

About 95% of what I do from a coding perspective, only requires Fable 5 Low and Medium. (and note, low and medium are out-performing Opus 4.8 and GPT 5.5 High and xHigh in a lot of benchmarks)

And I have been blown away with how good Fable Medium is at planning, it's better than Opus Max IMO. So thoughtful and detailed.

I have used Fable High on some really detailed architecture decisions, but have really focused on giving it tightly-scoped plans so I leverage the thinking depth correctly.

Fable has written over 1M lines of code for me in the last 24 hours.

And just to repeat, I'm on the $100/mo plan, not hitting any usage limits, never moved me down to Opus, haven't hit a single guardrail.

I think the biggest mistake ppl are making with this model is running it with High effort, in cases where Low or Medium will return the exact same output.

Still a lot more testing and benchmarking to do, but my guess is most engineers only need Fable High 1% - 5% of the time.

So if you're using High Effor 100% of the time, and hitting limits on your 20x plan, this is probably why.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@DeepakNesss https://t.co/Fw63wFAzo9

[Robin Ebers @robinebers]

bet you don't know this:

Fable 5 Low is CHEAPER / BETTER / FASTER than Opus 4.8 Max

if you use Cursor or similar and you pay per usage....

you'll save a ton of time and money + get better results

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@mitchellh What about composer 2.5 as coder instead?

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@Wotwot96 @LottoLabs This.

Pure LLM vibing just leads to spaghetti code down the line. Hard to debug.

Augmented coding is where it’s at. Hardest part of code isn’t the syntax but the systems thinking to create smth.

Like a house.

That’s why architects exist.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@seraleev If price isn’t a problem, a fully speced out Mac mini. That 516 ram is crazy for local ai

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@ImAnnoying_AF @SovereignIM Mhmmm, you have no idea how many phrases just stick in your mind for years.

Video, audio and images are dangerous tools.

For good or evil.

Just text is fine, tho.
I like Twitter & Substack.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@SovereignIM And images.

Just text is within our capacity, tho. I’d rather not be uninformed.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@SovereignIM Reading about it should be fine.

Similar to reading a report from a messenger about a city countries away.

It’s the video news and audio that’s not okay. Our mind can’t differentiate between reality and digital equivalent.

[𝐎𝐌𝐀𝐒𝐔 @MrOmasu]

@theralkia Probably rebels often.