Enchanted LLM Reviews

As the title suggests, you can’t submit anything after typing it in.

Kept crashing the server. It was a docker container but the whole server crashed

Was excited to see if this would work on my self-hosted llama instance. Unfortunately, it seems that responses are cut off and not fully displaying in the UI.

The UI in this looks beautiful, but it’s got a serious problem in that the button that you have to use to submit a prompt is located directly next to the record audio button and it seems like the record audio button is above it and there’s no way to actually hit the button to submit a request.

EnchantedLLM freezes :(

I assume Devs did not run any IOS18 betas to see EnchantedLLM crashing immediately when opened after installing this update. Some bugs here and there like text left in input field after submit and a response received, doesn’t happen all of the time though.

EnchantedLLM is jaw dropping, especially for a FOSS project!

EnchantedLLM somehow performs better and faster on my phone over 5G than my locally ran Open WebUI instance!

Plus it also supports multimodal models with the ability to attach files and images.

THE NEGATIVE:
Clients don’t sync across devices. This one limitation is the only reason I can’t daily drive EnchantedLLM. Please add the ability to sync the clients somehow.

I know security is a big deal hence the whole reason for self hosting the LLM to begin with. But iCloud has e2e encryption now (not on by default), WebDAV is an option. If you add File-provider support then options like Mobius Sync (Syncthing) and other apps would be available.

But simply put I use multiple devices throughout the day, and no syncing is a show stopper for me.

This is a really nice app that I’ve been using quite a bit. One suggestion, though: it would be nice if autocorrect was enabled. It’s surprisingly hard to type out a chat on an iPhone without that.

Unable to display formula symbols. Hope to optimize this.

Using llama3 on mac is so great that I even gave up chatgpt3.5.

i set a local ollama on my pc，now i can use EnchantedLLM instead of web page, it's very useful.

Works well with great features. How to use with Jan.ai?

Local LLMs have made significant progresses over the last months. I was looking for an app to help me ditch teh expensive online GPTs for something private and less censored (and also that stays consistent over time, not like GPT4...) EnchantedLLM was exactly what I was looking for.

clean, simple, works and free. thanks!!!

Great start for this new app. Has been working flawlessly w/ Ollama.

One of the few iPhone apps that is actually good

Works just as expected. If you’re looking for an chatbot app to replace your terminal, this is very darn near perfect.

EnchantedLLM connects to my laptop running ollama easily, great work!

Self hosted and privacy oriented

I have been using Ollama running on a pretty high powered machine, and I find it very nice to be able to interact with that from my other computers for my daily work (and even from my phone if I am burning off some idle time).

I wanted to access my local models on ollama and thought I would have to make my own app. Thankfully this exists and it’s nice, simple, and clean! I can connect to my models over Tailscale and have full privacy and control over my LLMs

Edit:
Only things I would like to see added:
- Stop scrolling to bottom when generating so you can read
- store multiple Ollama endpoints to select from