Stability AI, Thorn, and All Tech Is Human partner to prioritize child safety in AI development. They adopt Safety by Design principles to prevent AI misuse for child sexual abuse material and sexual harm. Annual progress updates will be released starting 2025.
AI | Rating: 42 | 2024-04-23 01:05:07 PM |
Stable Diffusion 3 and Stable Diffusion 3 Turbo are now available on the Stability AI Developer Platform API. The model outperforms DALL-E 3 and Midjourney v6 in typography and prompt adherence, based on human preference evaluations. The new Multimodal Diffusion Transformer (MMDiT) architecture uses separate sets of weights for image and language.
Rating: 87 | 2024-09-29 03:30:39 AM |
2024-04-16 04:27:52 PM |
2024-04-15 03:47:53 PM |
2024-04-12 07:28:16 PM |
MedARC, in collaboration with researchers from Princeton Neuroscience Institute, University of Minnesota, University of Sydney, and University of Waterloo, has released a new fMRI-to-image reconstruction model that works well with just 1 hour of training data.
Rating: 85 | 2024-09-29 03:30:39 AM |
2024-04-08 08:07:51 PM |
Stable LM 2 12B is a pair of powerful 12 billion parameter language models trained on multilingual data in English, Spanish, German, Italian, French, Portuguese, and Dutch. The models are available for testing on Hugging Face and can be utilized commercially and non-commercially with a Stability AI Membership. This release includes an update to Stable LM 2 1.6B.
Rating: 87 | 2024-09-29 03:30:39 AM |
2024-04-03 12:10:54 PM |
Stable Audio 2.0 is a new model that generates high-quality, full tracks up to three minutes long at 44.1 KHz stereo from a single natural language prompt. It also includes audio-to-audio capabilities, allowing users to upload audio samples and transform them into various sounds. The update also expands sound effect generation and style transfer.
Rating: 86 | 2024-09-29 03:30:39 AM |