SDXL 0.9 is a generative model recently released by Stability.AI and described in the report "SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis" by Podell et al. in July 2023.
Timestamps:
00:00 - SDXL 0.9
00:31 - Improving Latent Diffusion Models for High-Resolution Image Synthesis
01:38 - SDXL architecture
01:57 - Architecture and Scale
02:37 - Conditioning the model on image size
04:06 - Conditioning the model on cropping parameters
04:55 - Multi-aspect training
05:25 - Improved autoencoder
06:22 - Refinement stage
06:55 - Comparison vs Midjourney v5.1
07:58 - Limitations
Link to technical report: https://github.com/Stability-AI/gener...
Topics: #stablediffusion #ai #sdxl #stability
For related content:
Twitter: / samuelalbanie
Research lab: https://caml-lab.com/
personal webpage: https://samuelalbanie.com/
YouTube: / @samuelalbanie1
TikTok: / samuelalbanie
Instagram: / samuelalbanie
LinkedIn: / samuel-albanie
Discord server for filtir: / discord
(Optional) if you'd like to support the channel:
https://www.buymeacoffee.com/samuelal...
/ samuel_albanie