Upscales have a new button now called "Vary (Region)". This let's you reroll specific areas of the image. You can also type /settings & click "Remix Mode" to set new prompts over selected regions. Add a hat, remove a building, but as always have fun!
MULTIMODAL AI
-
Meta’s SeamlessM4T Models Now Available on Hugging Face
By
–
The new SeamlessM4T models from @MetaAI are now available on Hugging Face! https://
huggingface.co/models?search=
facebook/seamless-m4t
… -
IDEFICS Model Playground Now Available on Hugging Face
By
–
Straight from the new IDEFICS model playground https://
huggingface.co/spaces/Hugging
FaceM4/idefics_playground
…
h/t @BrigitteTousi -
SeamlessM4T: Single System Approach for Superior Translation Quality
By
–
Compared to cascaded approaches, SeamlessM4T's single system approach reduces errors & delays, increasing translation efficiency & quality, delivering state-of-the-art results. Want to see it for yourself, try the demo https://
bit.ly/3qEqehU -
Meta Releases SeamlessM4T Translation Model Openly
By
–
We believe SeamlessM4T represents a significant breakthrough and as part of our open approach, today we're publicly releasing this work under a CC BY-NC 4.0 license so that others can continue to build on this important field of study. Get the code https://
bit.ly/3E4mtVZ -
SeamlessM4T: All-in-One Multilingual Multimodal Translation Model
By
–
Introducing SeamlessM4T, the first all-in-one, multilingual multimodal translation model. This single model can perform tasks across speech-to-text, speech-to-speech, text-to-text translation & speech recognition for up to 100 languages depending on the task. Details
-
Fine-tuning and ControlNet Advances for SDXL Text-to-Image AI
By
–
Seeing the fast follow from the community racing to get fine-tuning and ControlNet working for SDXL has been exciting. These techniques bring text-to-image AI to life. Here are some examples of my dog Queso of SD2.1 Dreambooth (left) vs. SDXL LoRA (right). The quality is
-
Stable Diffusion 2: The Awkward Second Album
By
–
In October 2022, Stable Diffusion 2 was released. It was sort of like the awkward 2nd album. Good, just… different. The migration to OpenCLIP as the text encoder significantly changed image composition and limited the use of artists' names that could be used in prompts.
-
Stable Diffusion 1.4 Celebrates One Year Anniversary
By
–
On August 22, 2022 (1yr ago today!) Stable Diffusion 1.4 made its debut. Happy Birthday Stable Diffusion! At last, it seemed like it was finally possible to generate consistently good images from a text prompt.
-
VQGAN and CLIP Notebooks: A Significant Step Forward in Image Generation
By
–
In April 2022, @RiversHaveWings shared a series of Colab notebooks that combined VQGAN and CLIP. https://
x.com/RiversHaveWing
s/status/1516582795438567424
… This was a significant step forward. Images were starting to resemble their prompts, and textures like brush strokes and pencil marks were starting to emerge.