Which models were you using under this? If you used the quality default as exists in the interface, it makes sense that it was ~4x the cost as it'd be 3 frontier models judged by one of those.
The idea would be to use fusion with simpler, cheaper models.
A counterpoint to this is that we have some real different definitions of AI.
If you consider things like the machine learning filters in your smartphone camera and Google's AI Overviews for searches it's entirely plausible that the US is currently at 75%+ of AI usage.
I think it's your last point that's actually the strongest.
There's always gaps between theoretical and practical, but to see China investing so hard in the future while the US digs in it's heels is infuriating.
The price of energy sets a floor on the price of all manufactured goods. By kneecapping the cheapest sources of energy, the regime kneecaps all domestic manufacturers.
China's aggressive buildout of cost effective energy production isn't because they're 'woke,' it's because it makes them more competitive. Every product they export at low prices is in part due to the their extremely cheap energy.
It's like the regime looked at the UK's collapsing manufacturing industry due to their high energy costs and said "I want that for us!"
But hands down the most useful thing I've made is HutchDB, which is a MCP service that you can call from any AI chat or Agent setup to store data for you.
Literally from your AI you just say "save that to Hutch" and then it figures out:
- The schema + fields
- Builds nice webviews (Kanban, Timeline, Grid, Calendar)
- Lets you share the output with people
So people use it for all kinds of things like time tracking ("every hour save a summary of my activities to Hutch"), for Agent to Human handoff ("Every day check social media for mentions of my company and save them to Hutch").
I use it for things like recording all of our marketing activities and then having my AI compare those to signups for rough attribution, etc.
I love it! A few questions that made me think:
- how do you know that "people use it for all kinds of things"? I just read your privacy policy and I'm concerned about my AI agent possibly leaking some API key to Hutch and then you can read it.
- how is this free? You're hosting on Vercel, one of the most expensive hosting providers. What happens when this goes semi-viral? How do I know you won't just pull the plug to cut costs or start charging $500/month? I don't want to sign up, invest my time, and then lose access.
- I signed up and now I can't access privacy policy nor terms of service pages, because when I go to https://hutchdb.com/ I get redirected to https://hutchdb.com/dashboard
This is all very fair criticisms. This thread asked: "What are tools you have made for yourself?" and that's genuinely what this is. I wanted it so I made it and then AI makes it so easy to just throw up a marketing page.
I've a a handful of dev friends that have started to use it as well and give their feedback and it's been slowly growing as I've added sharing/invites.
I would absolutely not recommend putting big production data into it currently.
My vision for it was something more like how the #1 use of spreadsheets is actually people making lists and not actually people doing lots of calculations.
Given the uptake today (thanks everybody!) and your feedback (thanks Mystery-Machine) I'm going to work at addressing your concerns.
I thought it was more implied, but let me be more explicit:
- This is something I made for myself without a lot of commercial thought, so I still haven't thought through pricing + usage + limits + operational limits. In it's current wildly unoptimized state it's still very cheap to run.
- For the specific concern about API Key leakage there's not a lot I can do about that (that I'm aware of) as the logic of what gets sent is handled by the client AI. It is possible to pull down and audit both the tools + instructions that are published by the MCP server if there are concerns on that side.
It's not just comparing all the models, it's also comparing all the providers and configurations of those models.
If you're doing any kind of production AI work you'll end up with outages caused by calling a single provider, OpenRouter seamlessly switching between providers is a godsend for uptime.
But even more than that there's meaningful cost+speed differences.
Here's Sonnet 4.6 being served direct, via Amazon and via Google
WRT the native grammar, consider adjective order. Few native english speakers (me included) can off the cuff name the proper order, but everyone knows the "right" order.
We're already seeing other states struggling with soaring house insurance rates (California and Florida) and a pattern of spikes in rates leading government mandated caps on rates leading to insurers pulling out of the states.
I've no idea what the going rate for insurance is currently in New Orleans but it has to be crazy right?
I was curious just how much of a difference there was, so ran a quick eval comparing them and fwiw DeepSeek is considerably slower but much much ~5x cheaper than Haiku and fwiw ~35x cheaper than Claude Opus 4.7.
> I wouldn't be surprised if programmers had, collectively, written more game engines than actual games.
Tailored to web developers, there definitively are more half-finished frameworks sitting on people's disks than finished web applications, I'm sure my ratio is pretty close to 1/1 over the years.
Oh, that's exactly what I do. My rule is: one game, one engine. It's based on whatever the OS provides of course, or an abstraction layer like SDL, but everything above that is my own, tailored specifically to the game at hand.
Or when I wanted to write a novel and went into world-building fantasy enciclopedia for two years... I didn't even pass the page 2 of the novel, lol. Now it's all forgotten.
But how am I supposed to be productive writing blogposts unless I can copy my favorite Clojure templating library into Nix first, so I can have completely statically and reproducible blog posts building from markdown together with the nicest type of templating?
If we're all being honest, I'd rather read the Clojure/Nix templating blogpost instead of the 10,000th "why human interaction is important" bearblog essay.
yeah, part of my current writing push was made more successful by two things:
* I am not allowed to use a blogging system I wrote. (Really, I've written three or four at this point and need to stop, and there are plenty of existing systems that still align with my idiosyncratic constraints.)
* The blog must not have any meta content about blog tooling.
(I cheated a little on the latter by having an extra "site" blog for that - which lets me get the words out but doesn't "count" for the writing goal. A useful outlet, but it meant an extra month or so before "real writing" outnumbered meta writing :-)
I think it is great to combine two personal projects into one!
For me, I can't learn anything unless I actually have a purpose for it. So if I wanted to learn how to write a static site system, I would also need to think of a reason I need one!
one of my goals is to work on the server platform that i am using for my websites. i want to write a blog, but i am using that desire to push myself to work on the platform, so i need to complete that blog interface first.
I think writing a static site generator was the first moment I felt like I may be serious about this programming thing.
Those losers who still need Perl on their servers better be ready for a mind explosion
...thought, me back in (too lazy to look up which year it was). I probably published like two things with it, spent (what felt like) a million person hours on it, just to abandon it and use Textpattern.
But I am also currently writing version 2 of an app that utilizes a general-purpose application server that I wrote (took about seven months), at first. It's been shipping for a couple of years. The server does great, but is unwieldy. I suggest against writing general-purpose servers.
The new version uses a newer, more focused, server that is a lot lighter.
As expected, Fusion was 7x slower and 4x the cost.
This isn't a knock against it, just that it I think this places Fusion into a "use it only when you need it" category.
https://3fpi5avcqq.evvl.io/
reply