Google AI Ultra

I agree, and the problem is that "value" != "utilization".

It costs the provider the same whether the user is asking for advice on changing a recipe or building a comprehensive project plan for a major software product - but the latter provides much more value than the former.

How can you extract an optimal price from the high-value use cases without making it prohibitively expensive for the low-value ones?

Worse, the "low-value" use cases likely influence public perception a great deal. If you drive the general public off your platform in an attempt to extract value from the professionals, your platform may never grow to the point that the professionals hear about it in the first place.

I feel prices will come down a lot for "viable" AI, not everyone needs the latest and greatest at rock-bottom prices. Assuming AGI is just a pipe-dream with LLMs as I suspect.

> When you see the delta between what's currently being burnt by OpenAI and what they bring home, the sweet point is going to be hard to find.

Moore's law should help as well, shouldn't it? GPUs will keep getting cheaper.

Unless the models also get more GPU hungry, but 2025-level performance, at least, shouldn't get more expensive.

Costs more than seats for Office 365, Salesforce and many productivity tools. I don't see management gleefully running to give access to whole departments. But then again, if you could drop headcount by just 1 on a team by giving it to the rest, you probably come out ahead.

It does look like it comes with a few other perks that would normally cost a bunch too, specifically, 30TB of Google drive storage

If they really want to win, they should make a competitor for O1-pro. It’s worth $200 to reduce LLM babysitting needs by %10.

Perhaps they're not interested in beating OpenAI in the business of selling $1 for $0.50.

New technology always starts off available to the elite and then slowly makes its way down to everyone. AI is no different.

Do you have a source for the $1500 number? Seems pretty high.

It's 6x that. The median is 2x: https://chatgpt.com/share/682ceb2a-b56c-800b-b49d-1a24c48709...

I wonder if there's an opportunity here to abstract away these subscription costs and offer a consistent interface and experience?

For example - what if someone were to start a company around a fork of LiteLLM? https://litellm.ai/

LiteLLM, out of the box, lets you create a number of virtual API keys. Each key can be assigned to a user or a team, and can be granted access to one or more models (and their associated keys). Models are configured globally, but can have an arbitrary number of "real" and "virtual" keys.

Then you could sell access to a host of primary providers - OpenAI, Google, Anthropic, Groq, Grok, etc. - through a single API endpoint and key. Users could switch between them by changing a line in a config file or choosing a model from a dropdown, depending on their interface.

Assuming you're able to build a reasonable userbase, presumably you could then contract directly with providers for wholesale API usage. Pricing would be tricky, as part of your value prop would be abstracting away marginal costs, but I strongly suspect that very few people are actually consuming the full API quotas on these $200+ plans. Those that are are likely to be working directly with the providers to reduce both cost and latency, too.

The other value you could offer is consistency. Your engineering team's core mission would be providing a consistent wrapper for all of these models - translating between OpenAI-compatible, Llama-style, and Claude-style APIs on the fly.

Is there already a company doing this? If not, do you think this is a good or bad idea?

This is even the case with Gemini:

The Gemini 2.5 Pro 05/06 release by Google’s own reported benchmarks was worse in 10/12 cases than the 3/25 version. Google re routed all traffic for the 3/25 checkpoint to the 05/06 version in the API.

I’m also unsure who needs all of these expanded quotas because the old Gemini subscription had higher quotas than I could ever anticipate using.

You can just surf between Gemini, DeepSeek, Qwen, etc. using them for free. I can't see paying for any AI subscription at this point as the free models out there are quite good and are updated every few months (at least).

I am willing to pay for up to 2 models at a time but I am constantly swapping subscriptions around. I think I'd started and cancelled GPT and Claude subscriptions at least 3-4 times each.

> If I get this then sooner or later OpenAI or Anthropic will be back on top.

The Gemini subscription is monthly, so not too much lock-in if you want to change later.

This 100%. Unless you are building a product around the latest models and absolutely must squeeze the latest available oomph, it's more advantageous to just wait a little bit.

We just signed up to spend $60+/month for every dev to have access to Copilot because the ROI is there. If $250/month save several hours per month for a person, it makes financial sense

Corps will likely negotiate bulk pricing and discounts, with extra layers of guarantees like "don't use and share our data" on top

"AI will make us X% more productive. 100%-X% of you are fired, the rest get a $250/month license".

I believe Imagen 4 and Veo 3 (the newest image/video models) and the "deep think" variant are for Ultra only. (Is it worth it? It's a different question.)

They put anything that makes sense. I don't know if including random movies makes sense.

They got Youtube Premium which is like 15$. 30TB of storage, a bit excessive and no equivalent but 20TB is around 100$ a month.

As a consumer it seems to me the low hanging fruit for these super-premium offerings is some substantial amount of API credits included every month. Back when API credits were a relatively new thing I used LLMs infrequently enough I just paid $5-10/mo for API credits and used a desktop UI to talk to ChatGPT.

Now they want $200, $250/mo which is borderline offensive, and you have to pay for any API use on top of that?

I can’t see a way that anyone would be able to give uncapped access to these models for a fixed price (unless you mean it’s scoped to your own use and not for company use? Even then, that’s still a risk to the provider.)

Putting API use into the monthly plans doesn't make a lot of business sense. The only people who would sign up specifically to use API requests on a monthly plan would be looking to have a lower overall bill, then they'd pay-per-request after that. It would be a net loss.

These subscriptions explicitly do not suck in your proprietary data, it's all laid out in their ToS.