~ai-venturebeat | Bookmarks (568)

Ai2 released Tülu 3, a model that makes fine-tuning open-source LLMs easier and get its performance closer to closed LLMs like GPT-4o.

LLaVA-o1 breaks down the answer into multiple reasoning components and uses inference-time scaling to optimize each stage.

Amazon strengthens its position in the AI race with an additional $4 billion investment in Anthropic, committing to custom silicon and cloud dominance to compete with Microsoft and Google.

Lightricks releases LTXV, a breakthrough open-source AI model that generates professional-quality videos in real-time on consumer hardware, challenging tech giants with faster processing speeds and free access for developers.

We dive into the most significant takeaways from Microsoft Ignite, and Microsoft's emerging leadership in the area of AI agents.

Salesforce announced Testing Center, a platform that lets enterprises test AI agents responses without touching sensitive information.

AI agents need to consistent, and H2O.ai believes a mix of generative and predictive AI models is the answer.

VentureBeat's viewing of the AI Agent Space revealed only 19 distinct agent models available for use, a fry cry from cloud rivals.

Fable built its AI Sim Francisco "war game" to find out why OpenAI's behind closed doors board fight turned out the way it did.

With this chip, Ubitium hopes it will be able to replace dedicated processors for specialized computing tasks at the edge, like NPUs for AI.

The Chips & Science Act was a bipartisan law passed to give $52 billion to the U.S. semiconductor industry. Will Republicans stand behind it?

Wordware, a San Francisco startup, raised $30 million to simplify AI development with a natural language platform, enabling non-coders to build AI agents without traditional programming.

Data quality vendor raises new money as it expands its unstructured data quality monitoring technology in a bid to help accelerate enterprise AI deployment.

It enables AI agents to achieve a 98% success rate in multi-step tasks, compared to just 24% for agents using traditional methods.

OpenScholar, an innovative AI system by Allen Institute for AI and University of Washington, revolutionizes scientific research by processing 45 million papers instantly, offering researchers citation-backed answers and challenging proprietary AI systems.

Claude can perform impressively complex tasks, but it will also make stupid mistakes from time to time.

The company has partnered with Anthropic to bring Claude 3.5 family of models to Cortex AI, the fully-managed service for gen AI development

The company’s published results highlight its ability to handle a wide range of tasks, from complex mathematics to logic-based scenarios.

Gary Shapiro, the CEO of the CTA, is blunt when he articulates his views of politics, tech, antitrust law and economic growth.

For applications that do not require masking, the model can be configured to simply tag sensitive entities, providing options.

Bringing in more AI agents to workflows means having a strong orchestrator agent to manage it all for enterprises.

Adobe's SlimLM, a breakthrough AI system, brings advanced document processing directly to smartphones without the need for cloud computing, offering enhanced privacy and reduced costs for businesses.