The new issue of Synthetic Work is out, people.

This week’s free edition is titled Issue #1 – Burn the books, ban AI. Screwed teachers: Middle Ages are so sexy.

Read it here: https://t.co/USOaNd934y https://t.co/KFWRFBDsrn

The new issue of Synthetic Work is out, people.

This week’s free edition is titled Issue #3 – I will edit and humanize your AI content

Read it here: https://t.co/usXcjqWuJS https://t.co/OBvNuuMMvp

Who wants to try the new large language model by Meta called #LLaMA? I have the 65B parameters up and running and I’d be happy to run some prompts for you, if you are curious.

(remember that this is not a chatbot, so need to submit a phrase that the #AI will complete for you)

A *lot* of people showed interest in the synthetic pictures I recently generated with #stablediffusion, so I’m opening a few new sections of Synthetic Work dedicated to the output of AI-assisted labor.

Let’s start with:
https://t.co/Y5xCSPXA89 https://t.co/WFdLnavKXf

Now, in this section, there are only my pictures, but I hope that this will become a place where I can showcase the most amazing works generated by the #AI community.

If you are interested in appearing here, send me a message.

Another great release from @camenduru. If you want to see what Stable Diffusion and AUTOMATIC1111 can do, these colabs remove all the friction from trying.

I’m busy testing LLaMA this weekend, but this is in my list for next weekend. https://t.co/aggaNNXZV8

@jankais3r I’m reading now, thank you. Fine, then. Let’s focus on the “autocomplete” scenarios.

@Vote_ForPedro @ggerganov @jankais3r The word you are looking for is exhilarating 🙂

I think we have seen nothing, yet. I’m expecting @StabilityAI, directly or through the Open Assistant project, to go well beyond this. BTW, the 1st Open Assistant models are being trained as we speak, as I read on their Discord.

@jacobi_torsten @simonw The reason why I bought this high-end machine is that I was expecting to run large language models in the very near future. I had to wait just 1 month for it to happen.

Renting GPU online wouldn’t work for me as I do 100K generations with Stable Diffusion and TTS models.

@ggerganov @jankais3r Weird things that LLaMA 65B says: https://t.co/Hio2AyKBQV