On Wednesday, Galileo launched an Agent Leaderboard on Hugging Face, an open-source AI platform where users can build, train, access, and deploy AI models. The leaderboard is meant to help people ...

Integrated graphics cards have been fighting an uphill battle for many years, often failing to achieve anything near what ...

Researchers used questions from the NPR Sunday Puzzle challenge to build a benchmark to test AI 'reasoning' models.

How Goodhart’s Law Reveals the Opportunity in Long-Term Innovation Investing, and why traditional performance metrics may be ...

OpenThinker-32B achieved benchmark-beating results using just 14% of the data its Chinese competitor needed, marking a win ...

Strix Halo, AMD's upcoming and extremely large APU, has finally seen some benchmarks in 3DMark Time Spy. These early results ...

New report reveals one-third of service queries are solvable without a professional’s help and can be solved through self-serviceNEW YORK, Feb. 13, 2025 (GLOBE NEWSWIRE) -- Aquant, an AI platform ...

Using over 520 carefully designed questions, the benchmark assesses how well AI models handle both factual ... but showed weaker performance in reasoning-heavy bias questions.

Discover the strengths and weaknesses of o3-mini and DeepSeek R1 in this detailed AI model comparison of its coding skills ...

Just days after DeepSeek R1 made headlines, Moonshot AI introduced Kimi AI 1.5, a model already touted superior to OpenAI’s ...