Nvidia has launched Llama-3.1-Nemotron-Ultra-253B, a fully open-source language model that outperforms the larger DeepSeek R1 on several benchmarks despite having less than half the parameters. Carl Franzen of VentureBeat reports the model is now available on Hugging Face with open weights and training data. The 253-billion parameter model features a unique toggle for “reasoning on” …
Meta has released its newest generation of artificial intelligence models, Llama 4, introducing three variants with improved capabilities. The weekend release included two immediate offerings – Llama 4 Scout and Llama 4 Maverick – with a third model, Llama 4 Behemoth, still in development. According to Meta, Llama 4 models mark “the beginning of a …
Oumi has released HallOumi, an open-source claim verification model designed to detect and prevent AI hallucinations. According to Sean Michael Kerner of VentureBeat, the tool analyzes AI-generated content on a sentence-by-sentence basis, providing confidence scores, specific citations, and human-readable explanations. Led by ex-Apple and Google engineers, Oumi aims to build an unconditionally open-source AI platform. …
Researchers from Sentient, the University of Washington, Princeton University, and UC Berkeley have introduced Open Deep Search (ODS), a new open-source framework designed to match the capabilities of proprietary AI search solutions. The system combines reasoning agents with web search tools to enhance the performance of large language models. According to the research team led …
OpenAI intends to release its first “open” language model since GPT-2 in the coming months, according to a feedback form published on the company’s website. Kyle Wiggers reports that OpenAI is inviting developers, researchers, and community members to provide input on what they’d like to see in this new model. The company plans to host …
Meta’s Llama AI models have reached 1 billion downloads, according to CEO Mark Zuckerberg’s announcement on Threads. Kyle Wiggers reports that this represents a 53% increase from the 650 million downloads recorded in December 2024. The models, available under a proprietary license, power Meta’s AI assistant across Facebook, Instagram, and WhatsApp. Despite commercial restrictions in …
Google has announced Gemma 3, its latest collection of open-source AI models built with the same technology that powers its Gemini 2.0 models. Gemma 3 is specifically designed to run efficiently on a single GPU or TPU, making it accessible for developers working with limited hardware resources. The new model family comes in four sizes: …
Manus, an AI agent developed by Chinese startup Butterfly Effect, has generated significant buzz in the tech world. Described as “the first general AI agent” capable of autonomously executing complex tasks, Manus has been hailed by some as China’s second “DeepSeek moment” – referring to the earlier breakthrough when Chinese AI model DeepSeek R1 outperformed …
Alibaba’s Qwen Team has introduced QwQ-32B, a new open-source language model that matches the performance of much larger models like DeepSeek-R1 despite having significantly fewer parameters. The 32-billion-parameter model, released under the Apache 2.0 license, leverages reinforcement learning (RL) to enhance reasoning capabilities for complex problem-solving tasks. Key features and capabilities QwQ-32B demonstrates impressive performance …
Cohere’s research division has launched Aya Vision, an open-weight vision model supporting 23 languages. According to Carl Franzen’s report in VentureBeat, the model comes in 8-billion and 32-billion parameter versions and can analyze images, generate text, and translate visual content. Aya Vision outperforms larger models like Llama 90B while requiring fewer computational resources. The model …