So cool! Imagine if it was local. Any similar localLLM projects out there?
Posted by Own-Potential-2308@reddit | LocalLLaMA | View on Reddit | 1 comments
https://youtu.be/FpSJX59L7N4?si=SYCl8STqFxZnwg7a
SM8085@reddit
Similar how?
I feel really good about my efforts to make the bot play 'Guess LLaMa?' Where it constructs the assets using StableDiffusion then starts guessing against the player. Right now it's using the text descriptions but a major TODO is to feed the images to Gemma3 vision capable models.
So the bot picked things like 'wearing a miniature sombrero', 'relaxing on a long', 'waterproof fur', I need to teach the bot that verbs are not StableDiffusion friendly.