Google has announced the release of Gemini 3.1 Pro, a tool designed specifically for tackling complex problems where simple ...

Google has launched Gemini 3.1 Pro, its latest AI model designed for complex reasoning tasks. With a benchmark score of 77.1% on ARC-AGI-2, Gemini 3.1 Pro doubles the reasoning performance of Gemini 3 ...

Google has rolled out Gemini 3.1 Pro, the latest update to its flagship AI ...

ARC-AGI 2 had been created when ARC-AGI 1 seemed all but saturated, but it appears that ARC-AGI 2 won’t remain unsolved for ...

Sonnet 4.6 adds adaptive thinking and browser task gains with 4x higher token use than Sonnet 4.5, budget planning changes by task type.

Anthropic has launched Claude Opus 4.6, bringing improvements to long-context reasoning, accuracy in coding, and extended ...

Scientists have developed a new type of artificial intelligence (AI) model that can reason differently from most large language models (LLMs) like ChatGPT, resulting in much better performance in key ...

Deepseek, a Chinese company, has introduced its Deepseek R1 model, attracting attention for its potential to rival OpenAI’s latest offerings. Reportedly outperforming OpenAI’s o1 Preview in benchmarks ...

The human brain is very good at solving complicated problems. One reason for that is that humans can break problems apart into manageable subtasks that are easy to solve one at a time. This allows us ...

Nvidia researchers developed dynamic memory sparsification (DMS), a technique that compresses the KV cache in large language models by up to 8x while maintaining reasoning accuracy — and it can be ...