Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: I made a self-hosted ChatGPT UI

github.com

Post reply on HN
Print view View original

31–40 of 54 posts

Re: Show HN: I made a self-hosted ChatGPT UI

#31

Post by nico »

Version467 wrote: post #29

I like it. The chat.openai.com frontend is very slow and frequently breaks, so I would consider using this. Have you considered adding different tts providers? It doesn't get better than elevenlabs right now, but they are also much more expensive than for example the azure neural voices.

> is very slow and frequently breaks

True, and the free version does it lot, almost on purpose.

The paid version is a lot faster and doesn’t break as often, but it still breaks (eg. For the last two days, the chat list on the sidebar disappeared and it showed a message saying “don’t worry, your chats will show up eventually”).

Re: Show HN: I made a self-hosted ChatGPT UI

#32

Post by tottenval »

Version467 wrote: post #29

I like it. The chat.openai.com frontend is very slow and frequently breaks, so I would consider using this. Have you considered adding different tts providers? It doesn't get better than elevenlabs right now, but they are also much more expensive than for example the azure neural voices.

Yes, I plan to add other providers soon, and native text-to-speech as well.

Re: Show HN: I made a self-hosted ChatGPT UI

#33

Post by monkmartinez »

I have tried this and many, many other ChatGPT frontends. I recently did a search for "chatgpt" on GitHub and filtered for frontends, but I was a bit disappointed with the results. Most of them seemed to be pretty similar and didn't offer anything new or unique.

I'm really interested in finding a frontend with LangChain integration that can switch between chat mode and doc mode or something along those lines. It would be great to have a more versatile tool for communication and collaboration.

Do any of you have any recommendations or know of any projects that fit this description?

Re: Show HN: I made a self-hosted ChatGPT UI

#35

Post by ra1231963 »

kkielhofner wrote: post #30
neilv wrote: post #2

> Chat with GPT is an open-source, unofficial ChatGPT app with extra features and more ways to customize your experience. It connects ChatGPT with ElevenLabs to give ChatGPT a realistic human voice. Looks like only GUI aspects of the UI are self-hosted, but that the text and speech aspects of the UI (and the bulk of the computation and IP) are provided by two SaaS services. Self-hosted (and some degree of open) ML mo…

It's somewhat ambiguous language - "self-hosted ChatGPT UI" could lead many to believe it's completely self-hosted. However, sophisticated readers familiar with ChatGPT will know the model and weights haven't been released and absent a leak/hack/release by OpenAI a completely self-hosted ChatGPT solution is impossible. Eventually we'll almost certainly see a "Completely self-hosted ChatGPT equivalent" (similar to Dal…

Ironically, this sounded like an answer from chat gpt.

But more to the point, a fully self hosted solution (llama), even running on a cellphone, is entirely believable. Look at some of the recent developments with llama.cpp and Stanford over the last week.

Re: Show HN: I made a self-hosted ChatGPT UI

#36

Post by tagyro »

Thank you!

I can't wait to test this! As other have mentioned, the "free" chat frontend is slow and the "Plus" one, not much better. Also, at $20/month, based on my usage, it's actually more expensive than using the API.

The last hurdle: as ChatGPT is not GDPR compliant, it would be really interesting/useful to find a way to "hide" the queries from openai and prevent the usage of your input in future training - basically, a self-hosted, non-leaking, chatGPT.

Re: Show HN: I made a self-hosted ChatGPT UI

#37

Post by CharlesW »

Apologies if this is so unrelated as to be off-topic, but I'm new to this and so my mental model is incomplete at best and completely wrong at worst. My question is:

How would one create a "domain expert" version of this? The idea would be to feed the model a bunch of specialized, domain-specific content, and then use an app like this as the UX for that.

Re: Show HN: I made a self-hosted ChatGPT UI

#38

Post by tottenval »

monkmartinez wrote: post #33

I have tried this and many, many other ChatGPT frontends. I recently did a search for "chatgpt" on GitHub and filtered for frontends, but I was a bit disappointed with the results. Most of them seemed to be pretty similar and didn't offer anything new or unique. I'm really interested in finding a frontend with LangChain integration that can switch between chat mode and doc mode or something along those lines. It woul…

What specific features would you like to see?

Re: Show HN: I made a self-hosted ChatGPT UI

#39

Post by humanistbot »

Is this allowed under OpenAI's ToS? I just don't want to connect my account and then get it banned.

Edit: It seems like it is just using the API instead of the web interface, and thus charging my account each time. I originally thought it was injecting into the free web interface. But is changing the system prompt going to get me banned?

Re: Show HN: I made a self-hosted ChatGPT UI

#40

Post by helloplanets »

CharlesW wrote: post #37

Apologies if this is so unrelated as to be off-topic, but I'm new to this and so my mental model is incomplete at best and completely wrong at worst. My question is: How would one create a "domain expert" version of this? The idea would be to feed the model a bunch of specialized, domain-specific content, and then use an app like this as the UX for that.

Either you can try it out with a longer system prompt, or wait until OpenAI releases a fine-tune API for the gpt-3.5-turbo model. The system prompts aren't designed to be very long, so the fine-tune is definitely what you'd be looking for. But it's only provided for the older models, so it's outdated at this point.

I guess you could also try to tack on an extra layer before the actual API call, and make your own system that includes key bits of info to the prompt from a more specific data set. But I'd guess at this rate of new releases from OpenAI, it might be a safe bet to wait the couple of weeks until they update the fine-tune API.

Post reply on HN