Mercury 2 introduces diffusion LLMs to text, delivering 10x faster speeds for AI agents and production workflows without sacrificing reasoning power.

The new AI model uses diffusion reasoning to generate 1,000 tokens per second; it runs about 5x faster than Haiku, speed limits are ...

Intellectual property theft is still a fundamental part of what makes these kinds of models work.

With reported 3x speed gains and limited degradation in output quality, the method targets one of the biggest pain points in production AI systems: latency at scale.

The traditional model of memory proposes that different types of long term memory are processed in separate brain modules. New research shows activation of these modules overlaps.

Presearch’s “Doppelgänger” is trying to help people discover adult creators rather than use nonconsensual deepfakes.

Forbes contributors publish independent expert analyses and insights. A former tech executive covering AI, XR and The Metaverse for Forbes. This voice experience is generated by AI. Learn more. This ...

If you’re a coder or someone who follows AI benchmarks for fun (hey, I won’t judge), this model will excite you tremendously. For everyone else, prepare to be underwhelmed—or rather, prepare to wait ...

French AI startup Mistral has weathered a rocky period of public questioning over the last year to emerge, now here in December 2025, with new, crowd-pleasing models for enterprise and indie ...

Google's DeepMind just released WeatherNext 2, a new version of its AI weather prediction model. The company promises that it "delivers more efficient, more accurate and higher-resolution global ...

A few days ago, Google finally explained why its best AI image generation model is called Nano Banana, confirming speculation that the moniker was just a placeholder that stuck after the model went ...