# Made a Frankenstein trying to make an all in one local ai app. now i'm not sure it's worth it. — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/5c1f1d440bce3d94

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-10-25 14:42 ingested 2026-10-26 02:40

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1xf45hd/made_a_frankenstein_trying_to_make_an_all_in_one/)

## retained text



```
I tried to solve the problems of multiple applications I actually use , but the thing is that efficiency of using those applications has pushed me to try creating an app. Hutash. But it became a Frankenstein of sorts.

1. Comfyui like system (local and free) but without the complicated nodes and dependency issues. Has a internal engine built ground up. 

2. Openart AI / magnific like studio experience but without the cost.

3. Pinokio like store where u can find all apps a but without the duplication of resources(local space is limited)

4. Simple installation of apps and models like windows app. We created a .hutash format to just declare and a model package can be formed. Not only models but model piples can be built within the app like higgsfeild. The architecture was made keeping this in mind. We have a subtitles app, podcast app (vertical apps) which shared model resources with a studio app. Once installed all apps can use them. May be like OLLAMA

5. Last but not least cli and mcp to execute these apps a and generations. To make automation or agentic use possible. 

6. Many others. 

That was my goal and I feel like I have pushed to my best of abilities in all the aspects. Realsed few versions. Yes they are issues. 

But the thing is now I don't know where to push it. We are a very small team. Couple of guys. Worse is we kept building it without a feedback loop. Even though I want the public to contribute it, there are four or five repositories spread out into multiple parts. 

now i'm stuck. i'm trying to solve all these things together, and it's taking too much time and resources. maintaining and developing something like this also feels like it would need a much bigger community.

i'm not sure if this is worth pursuing anymore, or if i'm trying to solve too many problems at once.

am i over engineering something that doesn't need to exist, or is there actually something useful here? what would you focus on if you were in my place?

I just want help directing our resources in a focused direction. as i wish this app would be a beneficial for all who want benefits of local ai without the complexity. 
```



document fetched 2026-10-26 00:38 · canonical https://reddit.com/r/LocalLLaMA/comments/1xf45hd/made_a_frankenstein_trying_to_make_an_all_in_one
