What sort of tokens per second are you seeing with your hardware? Mind sharing some notes on what you’re running there? Super curious!

I was pleasantly surprised by many models of the Deepseek family. Verbose, but in a good way? At least that was my experience. Love to see it mentioned here.

Have you had a chance to try ControlNet? There are some some really cool workflows with this.

Stable Diffusion + ControlNet gives you a sense of ‘manual’ control over your AUTOMATIC1111/web-ui workflow.

I quite like it, and I think there’s a ton of potential if you combine it with other processing techniques. Prompts are fickle things, there is no one-size fits all, but hopefully these help you adjust and tune your prompts as you play around with the toolset.

If you’re looking for models, civitai is a good place to try out new styles, checkpoints, LoRAs, and other downloadable content you can use to enhance your Diffusion Suite.

If you’re not sure what ControlNet is, try starting with this video here which goes over the workflow I personally use for some of my own projects.

In the case you’re looking to break into generative video content (based on similar stable diffusion web-ui workflows), you’ll want to check out TemporalKit + EbSynth (which is also detailed in the FOSAI Nexus)!

FWIW, it’s a new term I am trying to coin in FOSS communities (Free, Open-Source Software communities). It’s a spin off of ‘FOSS’, but for AI.

There’s literally nothing wrong with FOSS as an acronym, I just wanted to use one more focused in regards to AI tech to set the right expectations for everything shared in /c/FOSAI

I felt it was a term worth coining given the varied requirements and dependancies AI/LLMs tend to have compared to typical FOSS stacks. Making this differentiation is important in some of the semantics these conversations carry.

Big brain moment.

Ironically, I think using this technology to do exactly that is one of its greatest strengths…

GL, HF!

Great suggestions! I’ve actually never interfaced with that first channel (SECourses). Looks like some solid tutorials. Definitely going to check that out. Thanks for sharing!

Lol, you had me in the first half not gonna lie. Well done, you almost fooled me!

Glad you had some fun! gpt4all is by far the easiest to get going with imo.

I suggest trying any of the GGML models if you haven’t already! They outperform almost every other model format at the moment.

If you’re looking for more models, TheBloke and KoboldAI are doing a ton for the community in this regard. Eric Hartford, too. Although TheBloke is typically the one who converts these into more accessible formats for the masses.

Thank you! Consider subscribing to /c/FOSAI if you want to stay in the loop with the latest and greatest news for AI.

This stuff is developing at breakneck speeds. Very excited to see what the landscape will look like for AI by the end of this year.

Absolutely! I’m having a blast launching /c/FOSAI over at Lemmy.world. I’ll do my best to consistently cross-post to everyone over here too!

Great suggestion! I’ll add that to the list.

For anyone unsure what this is, here’s a quick peek at Comfy-ui.

Kind of similar to ControlNet, it’s another method for expanding the options you have over your Stable Diffusion workflow.