Community Poll: Which LLM Provider Should AI Studio Automation Support Next?

I voted for a Google Gemini, not because I personally need it, but because I’m bullish on Google becoming one of the 2-3 major AI players in the long run. Currently, we are in an AI arms race, and I think we are still a few years away from those 2-3 major AI players emerging, likely after an AI-fueled stock market crash and some acquisitions. But there will likely only be 2-3 major players in the long run, maybe even just 2 (think Coke and Pepsi, Walmart and Target, Home Depot and Lowe’s, etc.).

To use a technology example of “major players”, take a look at the 1st screenshot below of web browser usage (source: ChatGPT). There are 2-3 major players. The others are minor players.

There are 2 reasons why I'm bullish on Google Gemini becoming one of the 2-3 major AI players in the long run:

1. Never bet against Google engineers. They continue to innovate. To illustrate that in the AI era, take a look at the 2nd screenshot below showing an excerpt from a November 25, 2025 article in The Economist titled Google has pierced Nvidia’s aura of invulnerability.

2. This is actually the bigger reason why I'm bullish on Google AI. Google owns YouTube. AI models have wowed us so far, but that's largely based on them “stealing the internet” and all of the human knowledge that it had built up over a couple of decades, largely in the form of blog posts and online forum posts. The AI models need a way to consume ongoing human insight, otherwise, they are just going to be “consuming their own AI exhaust”. Blog posts and online forum posts (with the exception of Reddit) are not as popular these days and I'll bet that most of them are now written by AI. My theory is that, moving forward, YouTube becomes the most popular way that human insight gets ingested by the digital world. YouTube is already the most popular distribution platform for creators. Since Google owns YouTube, Google gets exclusive access to use that content to continuously train their models with ongoing human insight.

So there's my overly wordy opinions on why I voted for Google Gemini 😀

I agree with Tim above, but I chose Gemini more fundamentally because we’re a Google Workspace shop.

My vote for Ollama is due to cost control and for ease of experimentation.

The use of a local, smaller, ‘less smart’ LLM for proving a concept with that means when switching to production and a ‘better’ LLM, it will almost certainly have the desired outcome.

​@TimRodmanTRAILD  - thank you for the insights! I agree with your sentiment.

Usually newcomers in the field come with an idea to disrupt a niche space. They are nimble and fast. 

However, Google plays the medium/longer-term game. Like a slow simmer. They have a whole suite of products that they have been investing in over years. Youtube, Chrome, Gmail, Sheets, Docs, Maps, Android, etc. And over that time they have been preaching about machine learning and artificial intelligence and research well before the current hype. Once they are ready and they launch a new model, it not only looks like it caught up with the competitors, but surpasses them in very impressive ways. They set a new bar. It seems like enabling AI capabilities for them in one go isn’t just a single new product for them but elevating and supercharging their ecosystem and its existing suite of products all at the same time.

Still, very interesting to see how everyone else is interpreting and experimenting with how they launch AI features in their products and how it resonates with the various user bases.

​@scottstilson - Are you currently using Gemini within your org outside of Acumatica? If yes, I would be very interested to learn how you are using it today (if you are willing to share)!

​@aiwan - thank you for sharing. So using Ollama almost as a stepping stone to another LLM. Are you currently using Ollama for other experimentation or business scenarios? I would love to know how!