HN Explorer

Show HN: A better UI for ChatGPT, Claude with text search, saved chats and more

23 points by youssefabdelm ↗ HN
Video introduction: https://www.youtube.com/watch?v=o9UwZcM62Dc

Link: https://www.teletyped.com/

Hey HN! My brother and I worked on this UI for ChatGPT, Claude, LLama 3, and Mistral models.

Our features:

- Full text search across all your chats

- Saved chats

- Switch models mid-chat

- Edit model outputs and continue generating, or save model responses for longer prompt engineering

- Temporary chats

- Edit model creativity (temperature)

- Different color themes

On our roadmap:

- Upload pdf files + photos (we know this an important one and plan to support in coming weeks)

- Web search + RAG for all models as a toggle

Really curious what you guys think!

21 comments

[ 4.5 ms ] story [ 65.5 ms ] thread
smcleod
The site looks pretty broken at the moment, the text appears to be half faded out / missing? https://i.imgur.com/iBQwr2f.png (FF125)
youssefabdelm
Do you mean the search text preview? That gradient is intentional and when you hover over it (inside the actual site), it brightens up.
spdustin
How does one hover when not using a mouse/trackpad?

I'll be more direct: it's unusable on touchscreen devices.

youssefabdelm
Good point, we could disable the fade on mobile.

The location of the gradient is parametric based on where the found keyword is. So if it's at the bottom, the gradient will be brighter there by default, or if the keyword is at the top, it'll be brighter there.

If there's more than one match in a single preview, the gradient accommodates for that as well. So if your keyword is "potato" and it shows up at the top and bottom, there is no gradient (or it's very small such that there is none)

There's typically enough text to discriminate (even before hover)

When the user hovers, the area that isn't bright gets brighter.

smcleod
Hmm, ok. It doesn’t seem that obvious, it just looks like the page never finished loading.
youssefabdelm
Still very useful to know, thank you for sharing. Will make the text brighter in general.
mutant
Absolutely unusable on mobile.
algoman
If you are referring to the scroll issue for the temperature dial, we just fixed that!
anonzzzies
Unlike almost everyone, I have some eye sight issues which makes me immediate switch off dark mode as it makes it harder to read text. On this site, it’s impossible for me read most of it.

Can’t please everyone, but this is quite rare.

rrr_oh_man
The text is in fact not very readable on iPhone mini (shifting yet always pretty low contrast, squished lines, that weird shadow/fade thing).

Might look pretty on desktop, but on mobile it’s just confusing.

youssefabdelm
Thanks for the feedback! Brightened the text for now on our site + landing page, but it's probably still too dark. We plan on supporting light mode soon and giving people that toggle, will keep improving in general too.
josefdlange
Just dropping in to say I appreciate the kind and friendly contextualization of your feedback. It would be really easy for someone in your position to be unfriendly in communicating this. This kind of tact and emotional intelligence is something I really cherish seeing from strangers on the Internet.
tarasglek
https://chatcraft.org is my open source project in similar vein. Please try it, we expose free models so one doesn't have to log in.

We have excellent mobile pwa, voice input, tts, search, sharing, make it easy to compare models.

We also do wasm to support running ts,go,ruby.

Its basically emacs disguised as a chat app. We wrote it as a way to utilize every cool feature llms have for own productivity, eg I can trivially build an adhoc clickhouse exploring tool in it. Can post examples if people are curious of specific abilities

hu3
Please post examples. I'm looking for such tool.
saasxyz
the site needs more working currently unusable but kudos to your design efforts
EarthAmbassador
Sorry, but very little is worth $26/month, in my view. If it operated as an uncensored model with guaranteed privacy, that might be worth something When combined with other features. Unfettered private creation and knowledge generation is a proper goal with LLMs.
youssefabdelm
Thanks for the feedback! We reduced the pricing to $20/month. I agree those features would be cooler, will definitely explore
saberience
It’s too easy to make a tool like this to charge that much.

Several other people have posted similar tools to Hackernews and I made my own personal one in about a week.

So yeah, ironically, gpts have made it way easier to make a basic gui to call some apis :)