Nano Banana Pro + Veo 3.1 is all you need.pic.twitter.com/rxz9ImBD7m
— Shubham Saboo (@Saboo_Shubham_) 21 novembre 2025
Nano Banana Pro + Veo 3.1 is all you need.
By
–
Nano Banana Pro + Veo 3.1 is all you need.pic.twitter.com/rxz9ImBD7m
— Shubham Saboo (@Saboo_Shubham_) 21 novembre 2025
Nano Banana Pro + Veo 3.1 is all you need.
By
–
Gemini 3 is mindblowing, and I can't even keep up with how this is gonna turn out!
By
–
Hard to say, but I think it may be independent of MHA vs GQA. Also, Olmo 3 7B uses MHA, and 32B uses GQA.
By
–
Grok 4.1 Fast just arrived.
— God of Prompt (@godofprompt) 21 novembre 2025
> 93% agentic accuracy
> 2 million token context (wow)
> Insanely fast
And it’s free. https://t.co/KgM6SM8rUI pic.twitter.com/dY3CYAL8n0
Grok 4.1 Fast just arrived. > 93% agentic accuracy
> 2 million token context (wow)
> Insanely fast And it’s free.
By
–
In their Olmo 2 report they had an ablation study showing it reduces the loss spikes during training (but they also included QK norm, so it's hard to say how much of that reduction is due to QK norm and their post norm flavor).
Maybe best of both words is to do both like Gemma

By
–
Olmo models are always a highlight due to them being fully transparent and their nice, detailed technical reports. I am sure I'll talk more about the interesting training-related aspects from that 100-pager in the upcoming days and weeks.
In the meantime, here's the side-by-side
By
–
Unlike humans, every time LLMs get better at something they get worse at something else, and there’s no fix in sight.
By
–
AI 1.0: GOFAI
AI 2.0: LLMs
AI 3.0: AGI
By
–
Just started reading… thanks for canonicalizing the Olmo 3 spelling! (Rolls much easier off the keyboard than OLMo 2 where I was never quite sure which exact letters had to be capitalized :D)
By
–
It's available in paid preview via the Gemini API using the model string 'gemini-3-pro-image-preview'