I just got Oobabooga running for the first time with Llama-2, and have Automatic1111, and ComfyUI running for images. I am curious about ML too but I don’t know where this start with that one yet.
For the uninitiated, all of these tools are running offline open source (or mostly) models.
I too find it hard to use anything other than GPT-4. It’s still so much better than other options even if the model has felt majorly nerfed compared to earlier releases.
WizardLM 30B at 4 bits with the GGML version on Oobabooga runs almost as fast as Llama2 7B on just the GPU. I set it up with 10 threads on the CPU and ~20 layers on the GPU. That leaves plenty of room for a 4096 context with a batch size of 2048. I can even run a 2GB Stable Diffusion model at the same time with my 3080’s 16GBV.
Have you tried any of the larger models? I just ordered 64GB of ram. I also got kobold mostly working. I hope to use it to try Falcon 40. I really want to try a 70B model at 2-4 bit and see how its accuracy is.
I just tried it a few hours ago. Indeed, it is quite good. I knew it when a NSFW prompt test on an uncensored model generated a stable diffusion picture of a robot skeleton and a snarky reply. Like, yay we finally have a bight spot with this one.
I have oobabooga and automatic1111. I have some ideas of making an infinite rpg game where I store long term memory in excel files and make the LLM call python functions to find memory that relates to current situations, and using automatic1111 to generate images for the game. Something like a MUD. I’m sure other people have already figured it out but that’s what I’m daydreaming about rn
Watch this ~1hr long video when you get the chance. He’s using the stalkerware LLM, but he also describes how to use langchain to parse data like what you are wanting to do.
I have oobabooga and automatic1111. I have some ideas of making an infinite rpg game where I store long term memory in excel files and make the LLM call python functions to find memory that relates to current situations, and using automatic1111 to generate images for the game. Something like a MUD. I’m sure other people have already figured it out but that’s what I’m daydreaming about rn
Unfortunately LLaMA 2 is not FOSS. Meta claims it’s open source, but it’s while the source is available it’s definitely not free as in freedom. There are strings attached.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !technology@beehaw.org
A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.
Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.
I’m playing with Stable Diffusion currently. For text I’m still using GPT-4.
I too find it hard to use anything other than GPT-4. It’s still so much better than other options even if the model has felt majorly nerfed compared to earlier releases.
Check out Wizard 30B Uncensored. IMO it’s about as good as NerfedGPT 4… except free and private.
What hardware does it take to run a 30B?
I’m running it in GPT4All (CPU-based) with 64GB of RAM, and it runs pretty well. I’m not sure what you’d need if you were running it on GPU instead.
WizardLM 30B at 4 bits with the GGML version on Oobabooga runs almost as fast as Llama2 7B on just the GPU. I set it up with 10 threads on the CPU and ~20 layers on the GPU. That leaves plenty of room for a 4096 context with a batch size of 2048. I can even run a 2GB Stable Diffusion model at the same time with my 3080’s 16GBV.
Have you tried any of the larger models? I just ordered 64GB of ram. I also got kobold mostly working. I hope to use it to try Falcon 40. I really want to try a 70B model at 2-4 bit and see how its accuracy is.
I just tried it a few hours ago. Indeed, it is quite good. I knew it when a NSFW prompt test on an uncensored model generated a stable diffusion picture of a robot skeleton and a snarky reply. Like, yay we finally have a bight spot with this one.
gpt4all has some decent models that I believe are Free. There is a python CLI/library that works with it, and others, called
llmStable Diffusion and Musicgen.
I have oobabooga and automatic1111. I have some ideas of making an infinite rpg game where I store long term memory in excel files and make the LLM call python functions to find memory that relates to current situations, and using automatic1111 to generate images for the game. Something like a MUD. I’m sure other people have already figured it out but that’s what I’m daydreaming about rn
I think Silly Tavern + Silly Tavern Extras could achieve this, it uses ChromaDB for infinite context.
Interesting, I’ll take a look at this eventually.
Watch this ~1hr long video when you get the chance. He’s using the stalkerware LLM, but he also describes how to use langchain to parse data like what you are wanting to do.
https://piped.video/watch?v=dXxQ0LR-3Hg&t=772
Yeah, that’s the idea. Thanks for the video!
I have oobabooga and automatic1111. I have some ideas of making an infinite rpg game where I store long term memory in excel files and make the LLM call python functions to find memory that relates to current situations, and using automatic1111 to generate images for the game. Something like a MUD. I’m sure other people have already figured it out but that’s what I’m daydreaming about rn
You went with Excel files as your database??? Are you Satan?
No I am
Thanks Satan!
Unfortunately LLaMA 2 is not FOSS. Meta claims it’s open source, but it’s while the source is available it’s definitely not free as in freedom. There are strings attached.
I’ve been playing with RWKV on my PC. Works pretty well and it’s 100% FOSS