OpenAI has released GPT-4.5 as a "Research Preview". The new language model is intended to be more natural and less hallucinatory, but is significantly more expensive than its predecessors. OpenAI has ...

Inception Labs has introduced Mercury, a new series of large language models that use diffusion technology rather than traditional autoregressive processing. The company reports these models can ...

OpenAI has also released a version of Advanced Voice powered by GPT-4o mini for all free ChatGPT users. According to the company, this feature offers conversation quality similar to the full GPT-4o ...

Elon Musk's AI company has released a voice feature for Grok 3 that allows explicit content and extreme personalities - in direct contrast to the approach of the competition. xAI has launched a new ...

Google has made its AI programming assistant Gemini Code Assist available for free to individual developers. The service offers up to 180,000 code completions per month and AI-powered code reviews on ...

Researchers from universities in China and Singapore, along with ByteDance, have created PhotoDoodle, an impressive new AI system for image editing. The model can learn different artistic styles from ...

Anthropic has taken a different approach from OpenAI with its latest release, introducing Claude 3.7 Sonnet as its first reasoning-enabled model that combines direct responses with extended thinking ...

After investing more than six months of development time and a year of GPU compute time, Hugging Face has published a free, open-source manual that provides detailed instructions for efficiently ...

Babuschkin claims on X that the prompt changes censoring Grok 3 were made by a former OpenAI employee "that hasn't fully absorbed xAI's culture yet." The modification was reportedly missed during the ...

A research team from ELLIS Institute Tübingen, the University of Maryland, and Lawrence Livermore National Laboratory has developed a language model called "Huginn" that can deepen its reasoning ...

A study by researchers from Korea University, Upstage AI, and AIGEN Sciences has identified specialized components in large language models that handle time-dependent information. These "temporal ...

Google DeepMind has revealed the cost structure for its Veo 2 video generation model, available through its cloud API platform. Creating a video with Veo 2 costs $0.50 per second in the Vertex cloud ...