I'm not sure about you but I've become a lil skeptical...
Not that it's a bad thing. In fact, I have quite an open mind.
But when I read that "Gemini 4 Pro" is "Outdoing" *both* ;
Opus 5.5
&
GPT 6 Astra
In "early benchmark tests"
It raises my eyebrows a bit.
Why?
Well for one. Opus 5.5 is AMAZING .
Astra I have less experience with but
Sol 6.1 has been pretty awesome.
And my experience with Gemini is...
Mixed.
On ONE side, it 'pretty good' for running code,
'read only' tasks... And the 'banana flash' models
are pretty compelling for images.
So I guess it depends on what you would use it for.
Like, there's a difference if you just need images for a blog or substack
post and needing help thinking through, building, or debugging a
complicated software project.
That said.
I what I am NOT skeptical on is : the reported 3-D rendoring capabilities.
Considering it's already proven it's image capabilities. 3D doesn't seem
like that far of a stretch.
But I guess what I'm trying to say here, is ;
I feel that Claude and Codex are the industry standards.
And I don't see that changing anytime soon.