🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

4-bit StableLM 1.6B is even more impressive. Not many 1.6B models work well when quantized to 4-bit. Pre quantized model in the 🤗 MLX Community: huggingface.co/mlx-community/… thanks to Ashish Easily >100 tokens per second on an 8GB M2 Mini:

Awesome work by Tianlin Liu showing the advantages of the SoftMoE router!

How is next-token prediction capable of such intelligent behavior? I’m very excited to share our work, where we study the fractal structure of language. TLDR: thinking of next-token prediction in language as “word statistics” is a big oversimplification! arxiv.org/abs/2402.01825

We’ve released our full paper on on the Stable Audio model 💿 arXiv: arxiv.org/abs/2402.04825 Code: github.com/Stability-AI/s…… Metrics: github.com/Stability-AI/s… Demo: stability-ai.github.io/stable-audio-d…… 🧵

Super cool brand-new results on MoEs for deep RL led by Johan S. Obando 👍🏽 Ghada Sokar and Timon Willi with Clare Lyle Jesse Farebrother Jakob Foerster Gintare Karolina Dziugaite Doina Precup and Pablo Samuel Castro! Soft MoEs do offer some nice gains 🙌

Announcing Stable Diffusion 3, our most capable text-to-image model, utilizing a diffusion transformer architecture for greatly improved performance in multi-subject prompts, image quality, and spelling abilities. Today, we are opening the waitlist for early preview. This phase

Ramon is proposing some fascinating ideas for integrating AI into gaming, and the potential outcomes could be significant. It's exciting to witness these ideas beginning to take shape, congrats!

We release Stable Code Instruct 3B today. A super fast & light model that helps you code in the most popular programming languages such as Python, Javascript or SQL. Awesome work by the language team at Stability! See video below & try it out here: huggingface.co/spaces/stabili…

We just released our newest & most powerful language model: StableLM2 12B. Get large-size model performance while only requiring medium-size model memory & latency! Multilingual, function calling & tool usage, RAG-friendly, safety tuned. Great work by a great team 🚀check it out!

Hoy anunciamos StableLM2 12B, un modelo de lenguaje open-source en español muy potente y de coste asequible. Entrenado desde cero con datos públicos, permite el uso de RAG, funciones y tools externas. Excelente para resumir y extraer datos de cualquier texto. DM si te interesa!

In addition, today we also release an improved version of our (already SOTA) StableLM2 1.6B! This tiny & fast model now handles function calling & tool usage too. These models can play a role in larger LLM systems. If interested in small language models, definitely check it out!

the chat variant is the new top DPO model on RewardBench, congrats Stability AI, dethroning tulu

Awesome to see PaLI go open! Big congrats to Neil Houlsby & the team! 🚀

Really awesome and detailed report about how to clean and enhance large web datasets to train LLMs. Big kudos to Guilherme Penedo Hynek Kydlíček Loubna Ben Allal Anton Lozhkov Colin Raffel Leandro von Werra and Thomas Wolf for such an open, thorough and necessary work! 😍

Today, we’re thrilled to announce the open weights for Stable Diffusion 3 Medium, the latest and most advanced text-to-image AI model in our Stable Diffusion 3 series! This new release represents a major milestone in the evolution of generative AI and continues our commitment to

The first few months at Bioptimus have been a blast! Check out our first released model.

Fue un absoluto placer hablar con Jaime Novoa de kaden sobre la industria de la inteligencia artificial, sobre cómo funcionan y se crean los chatbots en términos sencillos, el rol de los datos y el presente y potencial futuro del ecosistema de IA español. open.spotify.com/episode/65w00m…