Selfhosted
A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don't control.
Rules:
-
Be civil: we're here to support and learn from one another. Insults won't be tolerated. Flame wars are frowned upon.
-
No spam posting.
-
Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it's not obvious why your post topic revolves around selfhosting, please include details to make it clear.
-
Don't duplicate the full text of your blog or github here. Just post the link for folks to click.
-
Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).
-
No trolling.
Resources:
- selfh.st Newsletter and index of selfhosted software and apps
- awesome-selfhosted software
- awesome-sysadmin resources
- Self-Hosted Podcast from Jupiter Broadcasting
Any issues on the community? Report it using the report flag.
Questions? DM the mods!
view the rest of the comments
The best/easiest way to get started with a self-hosted LLM is to check out this repo:
https://github.com/oobabooga/text-generation-webui
Its goal is to be the Automatic1111 of text generators, and it does a fair job at it.
A good model that's said to rival gpt-3.5 is the new Falcon model. The full sized version is too big to run on a single GPU, but the 7b version "only" needs about 16GB.
https://huggingface.co/tiiuae/falcon-7b
There's also the Wizard-uncensored model that is popular.
https://huggingface.co/ehartford/Wizard-Vicuna-13B-Uncensored
There are a ton of models out there with new ones popping up every day. You just need to search around. The oobabooga repo has a few models linked in the readme also.
Edit: there's also h20gpt, which seems really promising. I'm going to try it out in the next couple days.
https://github.com/h2oai/h2ogpt
Note that when using llama-derived models, such as vicuna, you are bound by their license to only use them for "research" purposes.
If you want an unrestricted version, go for open-llama or RedPajama.
Falcon is less restrictive and only wants a cut of profits if they exceed 1 million dollars, but I'd wager that fully unrestricted is the way to go.
Falcon has switched to Apache 2.0 and removed the commercial limit.
Sorry, I must've missed that somehow, then my comment only applies to llama and its direct derivates.
How do you know how much ram the model needs?
The model creator usually mentioned it in the readme:
Usually the models support CPU inference. Tremendously slow but works in a pinch.
There's an average correlation between the models parameters and the execution precision being used (eg. 7b parameters at f16 precision). And then using optimized execution for 8 bit or even 4 bit will reduce memory usage and increase execution time.
It's entirely dependent on the model, the framework, the hardware (CPU vs GPU).
Generally there should be some indication somewhere in the model's repo that states what you need.