Najnowsze wiadomości i analizy SI | Yellow

OpenAI's GPT-5.6 Sol and Anthropic's Claude Fable 5 split the coding crown, with Sol topping one agentic benchmark at 88.8% while Fable 5 holds the multi-file lead. Key Points: GPT-5.6 Sol leads Terminal-Bench 2.1 at 88.8%, ahead of Claude Fable 5's 83.4%. Fable 5 still tops SWE-Bench Pro at 80.3%,

Nous Research released Hermes Mixture of Agents 2.0 on Sunday, which combines outputs from multiple large language models, including GPT, Claude, and DeepSeek, to produce responses that outscore any individual model on standard benchmarks. According to a report, the MoA 2.0 is an update to Nous Rese

Google restricted Meta's access to its Gemini AI models around March, unable to supply the compute Meta wanted even as its cloud backlog swelled toward $460 billion. Key Points: Google told Meta around March it could not supply the full Gemini capacity it wanted to buy. The shortfall delayed several

A new OpenAI study finds its Codex agent now generates 99.8% of the output tokens the company's employees produce, signaling that autonomous AI has overtaken chatbots at work. Key Points: Codex produced 99.8% of output tokens among OpenAI staff, against 63.3% at outside organizations and 16.5% for i

Anthropic raised $65B at a $965B valuation, moving past OpenAI and tightening the contest for enterprise artificial intelligence. Key Points: Anthropic closed a $65B Series H round at a $965B post-money valuation. The deal puts the Claude maker above OpenAI’s earlier private-market benchmark. Cloud