Tag: llm

Deepset has successfully raised $30 million in funding to assist enterprises in leveraging the full potential of Large Language Models (LLMs). This investment will fuel the company's mission to make advanced AI accessible and practical for businesses seeking to integrate LLM capabilities into their operations.

Explore the limitations of LangChain and the rise of Atomic Agents as a more flexible and efficient paradigm for LLM development, offering granular control and enhanced modularity.

Microsoft introduces CollabLLM, a novel framework designed to enhance collaboration between Large Language Models (LLMs) and human users. This approach focuses on enabling LLMs to understand and adapt to user preferences and feedback, fostering a more intuitive and effective interaction. CollabLLM aims to bridge the gap between AI capabilities and user needs, paving the way for more sophisticated and personalized AI-assisted workflows.

This tutorial details how to leverage expert parallelism on Amazon SageMaker to accelerate the pre-training of the Mixtral 8x7B model, a powerful Mixture-of-Experts (MoE) large language model. We will guide you through the setup and configuration necessary to optimize distributed training for MoE architectures.

Magic AI introduces HashHop, a novel method for assessing Large Language Models' (LLMs) ability to process ultra-long contexts, offering a more robust alternative to existing techniques like Needle in a Haystack.

This guide provides an in-depth look at creating advanced round-robin multi-agent workflows using Microsoft AutoGen. Learn to orchestrate complex agent interactions for sophisticated task automation and problem-solving.

Explore AI2 Olmo, a groundbreaking system that enhances Large Language Model (LLM) traceability by linking generated content back to its source data. This tutorial provides an instructional overview of its capabilities and implications for AI transparency and research.

Apple is reportedly developing a ChatGPT-style application to significantly enhance Siri's capabilities, signaling a major AI investment aimed at competing with rivals like OpenAI and Google. This initiative, driven by a dedicated team, suggests a strategic shift towards more advanced conversational AI for its virtual assistant.

Apple’s internal ‘Veritas’ project is reportedly a significant, multi-year effort aimed at fundamentally revamping Siri. This initiative seeks to imbue the virtual assistant with advanced conversational abilities, a deeper understanding of context, and more natural interactions, addressing long-standing criticisms of Siri’s current limitations and positioning it to compete more effectively in the AI-driven landscape.

Microsoft has identified a novel phishing campaign leveraging Large Language Models (LLMs) to create sophisticated and evasive attacks, posing a significant challenge to traditional security measures. This analysis delves into the mechanics of these attacks and their implications.

Explore the top 44 Large Language Models (LLMs) defining the technological landscape in 2025. This report details their capabilities, applications, and the trajectory of AI innovation.

Geeky Gadgets introduces Gemini 2.5 Flash Lite, a powerful new AI model offering an unprecedented 1 million-token context window at an incredibly low price point of $0.40, democratizing access to advanced AI capabilities for developers and businesses.

This analysis delves into LAMEHUG, a novel malware leveraging Large Language Models (LLMs) for sophisticated, dynamic reconnaissance and data exfiltration, marking a significant escalation in cyber attack capabilities.

Microsoft Copilot Studio is expanding its AI model capabilities with the integration of Anthropic's Claude, offering users more choices and advanced AI features for building custom copilots.

Deepset has secured $30 million in funding to bolster its offerings in MLOps for Large Language Models (LLMs), aiming to enhance the development and deployment of AI applications.

This tutorial explores how to improve Retrieval-Augmented Generation (RAG) pipelines in Haystack by integrating DiversityRanker and LostInTheMiddleRanker, focusing on enhancing answer relevance and reducing information overload.