Jesus RodriguezinTowards AIInside DataGemma: Google DeepMind’s Initiative to Ground LLMs in Factual KnowledgeThe model comes accompanied by DataCommons, a data repository based on factual data.1d ago1d ago
Jesus RodriguezinTowards AIInside xLAM: Salesforce’s Models Specialized for Agentic TasksThe family of models is highly optimized for function calling.Sep 101Sep 101
Jesus RodriguezinTowards AIInside GameNGen: Google DeepMind’s New Model that can Simulate Entire 1993’s DOOM Game in Real TimeGameNGen represents a major milestone in creating generative AI models that can interact with complex real world environments.Sep 21Sep 21
Jesus RodriguezinTowards AIHow NVIDIA Pruned and Distilled Llama 3.1 to Create Minitron 4B and 8BThe new models are using state of the art pruning and distillation techniques.Aug 261Aug 261
Jesus RodriguezinTowards AIInside The AI Scientist: The AI Agent for Open-Ended Scientific DiscoveryThe framework combines different generative AI models to streamline scientific research from idea to paper.Aug 202Aug 202
Jesus RodriguezinTowards AIMeet PromptPoet: The New Prompt Engineering Framework that Everyone is Talking AboutOriginally created by Character.ai, PromptPoet abstracts some of the core building blocks of prompt engineering.Aug 121Aug 121
Jesus RodriguezinTowards AIMeet Gemma Scope and ShieldGemma: Google DeepMind’s New Releases for Interpretability and…The two frameworks are part of the Gemma 2 release.Aug 6Aug 6
Jesus RodriguezinTowards AIInside DeepMind’s AlphaProof and AlphaGeometry 2: Two Models that Achieved Silver Medal Status in…One model focuses on algebra and number theory, while the other mastered geometry.Jul 29Jul 29
Jesus RodriguezinTowards AIInside NuminaMath: The AI Model that Took The First Place In the AI Math OlympiadThe model used strong data curation, fine-tuning processes, and algorithmic improvements to reach the top of the AIMO leaderboard.Jul 22Jul 22
Jesus RodriguezUnderstanding FlashAttention-3: One of the Most Important Algortihms to Make Transformers FastThe new version takes full advatange of H100 capabilities to improve attention in transformer models.Jul 15Jul 15