Great work. This is neat kind of like age old there used to be hackintosh for macs. Now for ios. With this do you have access to app store. I think there is a unique identifier that android will lack for it to work.

> kind of like age old there used to be hackintosh for macs

Have you checked out the hackintosh community lately? They have a near fully functional port to Snapdragon now.

https://github.com/Snapintosh/Snapintosh

Hi. I built a free educational project that helps people explore how changes in DNA can affect genes and the body.

It includes guided examples and an optional in-browser assistant. It works in six languages, supports read-aloud, and requires no login.

History is repeating. I had heard while back myspace making comeback to I guess that is not happening. At the time it was real innovation to be able to stay connected to long lost friends who moved after college, etc. Then it got to extreme.

There is significant congregation of investment towards companies like this whereas many other with meaningful innovation are struggling to survive. The model is not a revolution like I must say how llm did at the face value. Yes, transformers were there for long but the use case etc. If you do cost comparison using regular model and setting it with prompt and structured output you can get result very close. This is what most open Jev and other are doing with tuning llm.

Wonder how would Fable 5.1 perform. I guess to expensive for such experiment. I guess you had Max 20x. Still it taking 3 hours is significant time for agent. What thinking level you had it on?

This is cool! Tried the tool. It seems good for images without much graphics.

I am glad the tool was useful to you; although it isn't perfectly precise, it gets the job done for anyone wanting to extract text quickly without downloading any software.

Couldn't agree more! The ones that are found out often have network that pushed their product or build. Plenty of silent builders make amazing apps, tools. Every once in a while you stumble upon think of packages libraries on npm, pip. Langchain is one example of how it took off.

This is neat. Thanks for sharing!

Haiku doesn't seem most cost effective solution. Jev like classifier can work in a fairly smaller model which are 1/10 of the cost. Opus is SOTA so I get the use case for one being restricted to Anthropic ecosystem. I think the use case for Haiku is mainly for users using on their chat for pro subscribers and free users to maximize their quota.

Good read! There is still blind application going on for the sake of using AI like blockchain when it came out. It is unfortunate many businesses only realize after they incur complaints or loss.