Constance Brown
178 posts
Honest opinion 👀
Do you guys like the new Antigravity 2.0
or do you still prefer the old IDE experience?
English
@DesignCntrl @Dr_Singularity No doubt, I understand where you’re coming from. Even if not in its current form. I do believe AI will be a transformative technology. We do have the world’s most brilliant minds working towards the goal. I’m optimistic.
English
@ConstanceB53pe @Dr_Singularity No one wants to pay $30/Mtok forever.
You can't do research at $30/Mtok.
English
this is pretty amazing news
Google DeepMind's AI agent autonomously solved 9 of 353 open Erdos problems.
Cost - only few hundred dollars per problem.
This is a major signal that AI is moving from helpful math assistant to actual autonomous research engine.
LLM agents connected to Lean are now solving real open mathematical problems, not just textbook exercises.
The strongest system resolved 9 open Erdős problems and proved 44 OEIS conjectures, with some breakthroughs costing only a few hundred dollars each.
Mathematics is one of the hardest domains because you cannot bluff your way through it. A proof either checks or it does not. And now AI systems are starting to generate proofs that survive formal verification.
1000x acceleration of progress is coming.
English
@DesignCntrl @Dr_Singularity That’s the timeline for AGI - I watched a video where he said that they’ve already processed all the protein with Alphafold. That’s just one part. The next part is supposed to take up to a decade.
English
@ConstanceB53pe @Dr_Singularity His timeline is still 5 years.
If AI was so good, you'd think we would use it now.
English
@Dr_Singularity only $105,900 to solve useless math problems instead of cancer.
English
@thealokverse @GeminiApp I honestly didn’t expect the to cook 👨🍳. Google is strong at a lot of things and in my opinion UI wasn’t one until now.
English
@EdenKollcinaku I prefer Gemini over Bard. I think Bard sounds old fashioned and random. Also, the story of Gemini came about from the merger of DeepMind and Google Brain is good for the lore.
English
70% of your marketing is the product name:
imagine if Google didn't change the name of "Bard" to "Gemini". it would sound cool, trendy and resonate more with gen z.
don't get me wrong! i love gemini models and all the google stuff.
but the word "Gemini" sounds like a long word. it has 3 vowels.
"Bard" on the other hand was simpler, cooler and more mysterious. it sounded more like a pro IT guy that'd help you with your tasks.
even OpenAI now is trying to promote the word "Codex" rather than "ChatGPT".
which name do you prefer most?
English
@BenittoJD @SaiNemani1 Thr only thing we’re sure about is that you know absolutely nothing about 3.5 pro. The end.
English
I am damn sure Google Gemini 3.5 Pro will be a flop. I can see in a flash what has happened. It is a benchmaxxed model.
I have been using the model, and to be frank, it screwed my codebase. It literally screwed my codebase. I thought this model would at least have some good design sense and taste, but it doesn't have that either. It doesn't have anything.
The only thing I will say is that it is fast, which is amazing. But what is the use of making it faster if it doesn't give the proper result?
Also, the model is expensive, man. Why is it so expensive? It is solving nothing. Composer 2.5 is a winner.
English
@simulx4 @th3mus1cman Give me something more that trust me bro. Are people gullible enough to just believe some random guy in the internet that he knows x without any proof. Actually, nvm.
English
@th3mus1cman bro nobody at Google uses Gemini models lol
I know dozens of kids who work at Google that use codex and Claude
Google wants them to use it and they actually try to make them. but everyone just copy pastas from better models
English
Just found out apparently 75% of code at Google is being written by AI now.
That scares me as someone who has used the Gemini models.
English
@juleshatton @_mohansolo Would be nice to keep at least the previous model but I’m assuming it has to do with compute.
English
@_mohansolo Please bring back Flash 3.0 in Antigravity - it was an excellent and efficient low-cost model.
English
Thanks for all of the Antigravity feedback over the last couple of days, especially around the IDE. Our intention was never to remove the IDE support for developers, and we should have been clearer with that in the product from the beginning.
We’ve made it clearer in 2.0 on how to connect to the IDE, fixed issues with opening the IDE on Windows machines, provided instructions to restore IDE settings & extensions, and more.
New releases for the Antigravity IDE and Antigravity 2.0 have rolled out with these changes.
We should have done better so we’re going to reset everyone’s Gemini quota for the week again.
English
@sharjeeltwitess @NishiVyx @grok 👀👀 who know what AI created these images. The one labeled Gemini wasn’t even created by Gemini.
English
@fragermk @rakyll @BuildurBrand_io @killedbygoogle While I don’t agree with this, it’s pretty funny 🤣🤣🤣
English
@rakyll @BuildurBrand_io @killedbygoogle Google’s 1 year plan: rip everything up and try again…
Google’s 2 year plan: rip everything up and try again…
Google’s 3 year plan: rip everything up and try again…
Google’s 4 year plan: rip everything up and try again…
Google’s 5 year plan: just sell ads…
English
I'm just going to say it.
Google sucks right now.
And they literally said why:
"Our product roadmaps are 120 days max - we haven't worked on a 1+ year roadmap in months."
This is from an executive, btw.
It feels like Google is throwing darts and praying something sticks.
English
@ChartsJournalX @LexnLin @OfficialLoganK As much as I like Google, I always take their benchmarks with a grain of salt. Spent a few hours yesterday testing 3.5 in the new antigravity cli. So far, I’m not impressed but still needs a lot more testing.
English
@ConstanceB53pe @LexnLin @OfficialLoganK They compare it themselves. Sounds too good to be true, to be honest.
x.com/Hesamation/sta…
ℏεsam@Hesamation
so Gemini 3.5 Flash is almost in the same intelligence level of Opus 4.7 while being ~3x cheaper. this is amazing, and also a reminder to always wait a few months when it comes to Gemini models and their benchmarks.
English
Gemini 3.5 feels like the start of a new era for Gemini, we spent the last 2.5 years putting the infrastructure, products, team, etc in place (learning lots of lessons along the way).
The model is the product, please keep the feedback coming!
English
@glproductions @LexnLin @OfficialLoganK I’m not really addressing trust just the comp. I daily drive Sonnet 4.6. I’m currently testing 3.5 in Antigravity Cli, I’m not impressed at all.
English
English
Gemini is way too slow to update. The cadence of OpenAI and Anthropic is faster. Deepmind has to be more nimble. Other parts of Google can follow-up with implementation but waiting to push a new model in all Google surfaces hurt.
Leon Lin@LexnLin
i do hope for the best @OfficialLoganK but - gemini updates take too long compared to openai/anthropic - features are not avaible - too many scattered products - It's just my own experience but I don't think Flash can now replace anything like Opus/GPT I hope that the pro model is going to be good!
English
@LexnLin @OfficialLoganK Why would you compare flash to Opus? It’s sonnet the comp ?
English
@OfficialLoganK i really really hope that 3.5 pro is going to be good because 3.5 flash feels is benchmaxxed. It's not at the same level of opus/gpt at all
English