EllyEleven

made you look

[EllyEleven @EllyEleven]

@antigravity Any usage improvements?

[EllyEleven @EllyEleven]

@OfficialLoganK @GoogleDeepMind Gemini 3.8 flash beats opus 5 on finance, legal, chart reasoning, video understanding.
That too at a fraction of price.

Gemini is now choosing their own fights to fight!

[EllyEleven @EllyEleven]

@GoogleAIStudio Wild contrast between terminal 2.1 and terminal bench 4
89.4 vs 19.1
Benchmarks really can tell different stories

[EllyEleven @EllyEleven]

@VadimStrizheus But 100 dollar plan, that's provided better per dollar value

[EllyEleven @EllyEleven]

How come this is efficient as per you, but everytime I use this fable 5.1 it consumes my usage after than any other model?
Is it a bug, or tell me what's efficient?

[Claude @claudeai]

We’re introducing Claude Fable 5.1 and Claude Mythos 5.1.

They're the world’s most advanced models for coding and knowledge work.

[EllyEleven @EllyEleven]

@antigravity It could be like ..go on try more!

[EllyEleven @EllyEleven]

@antigravity Renaming max thinking?

[EllyEleven @EllyEleven]

@OfficialLoganK gemini 3.8 flash is launched already?
I just saw that in antigravity,
Can you post its benchmarks??

[EllyEleven @EllyEleven]

@GoogleDeepMind And you brought Gemini 3.8 flash too..

[EllyEleven @EllyEleven]

@aisearchio I really thought opus 5 is bad until I used fable 5.1
I can't even use it properly to tell is it even good or bad.

It's only motive is to eat all my usage away!

[EllyEleven @EllyEleven]

Gemini 3.8 flash just dropped.

@GoogleDeepMind why are you not hyping it up??

[EllyEleven @EllyEleven]

@johnternus Let's get that m6 cheaper

[EllyEleven @EllyEleven]

@thsottiaux More than ever, now on astra

[EllyEleven @EllyEleven]

@samueljmcd It just blasts through usage like a hot knife through butter

[EllyEleven @EllyEleven]

@cursor_ai Cursor is not the ideal place to use such a pricey model!
But their harness is too good to ignore:(

[EllyEleven @EllyEleven]

@bcherny That's why you were reclusive to vague post!!
Because you had fable 5.1 to post.

[EllyEleven @EllyEleven]

@ClaudeDevs You should start giving the banked resets too!

The timezone and time difference between usages are too broad!

[EllyEleven @EllyEleven]

@OpenAI Astra shows up to be more efficient.
And we can get more usage per dollar then!

[EllyEleven @EllyEleven]

@claudeai With the fable 5.1 medium is more smarter than fable 5 max and it's even cheaper!!
So they did make models efficient now!!
And in top of this usage just resets!

[EllyEleven @EllyEleven]

@ClaudeDevs Communicating properly could have saved you a lot.
But this is not the right time to decrease usage!
Opus 5 is not loved that much,
Codex really gained 25 million users,
And with you decreasing the usage, people will jump boats.