AI - AI Dynamics

Comparing Stable Diffusion 1.5 vs 2.0 Image Generation Quality

By

–

25 November 2022 2h34

easy to compare a lot of images from both models on http://
stableboost.ai , e.g. "cute dog cooking tacos, photorrealistic", grid of boosted images from 1.5 (left) and 2.0 (right). 2.0 looking more distorted, cartoony, simpler, ignores text more. may need more prompt engineering

→ View original post on X — @karpathy

25 November 2022

Stable Diffusion 2.0 Shows Quality Decline Compared to 1.5

By

@karpathy

–

25 November 2022 2h34

plot twist: stable diffusion 2.0 looks quite a bit worse on the few prompts i've tried so far compared to 1.5 (even not including celebrities/artists). Running theory seems to be this is due to an aggressive data sanitization campaign since the original release (?).

→ View original post on X — @karpathy

25 November 2022

AGI-First Limited Edition T-Shirt NFT Airdrop Montreal

By

@montreal_ai

–

25 November 2022 2h28

"AGI-FIRST" T-Shirt By: http://
MONTREAL.AI Very Limited Quantity Available Now : https://
quebecartificialintelligence.com/shop Proof of Wear = Airdrop of the Limited Edition AGI-FIRST T-Shirt NFT #AGI #AGIFirst #MontrealAI

→ View original post on X — @montreal_ai

25 November 2022

Wealth Redistribution and Indigenous Rights: A 2023 Vision

By

@johnchavens

–

25 November 2022 1h08

2023 – the year that people start loving themselves enough to say, “hey, 50 richest people in the world – give up 90% of your wealth & save the world after recognizing and honoring the indigenous & marginalized people who have been ignored for decades!” It will be a good year.

→ View original post on X — @johnchavens

25 November 2022

Code Injection in Notebooks for Parameterization

By

@jeremyphoward

–

25 November 2022 0h03

Have you checked out the ability also to inject code into a cell to parameterise notebooks?

→ View original post on X — @jeremyphoward

25 November 2022

Current Limitations of Image Output Sizes and Scheduler Options

By

@replicate

–

24 November 2022 23h27

It doesn't yet support a few things. It only outputs 512 and 768 sized images, doesn't support init_image or mask, and it only has Euler and DDIM schedulers.

→ View original post on X — @replicate

24 November 2022