Anthropic's new 'Mythos' models reportedly surpass Claude Opus 4.6 in coding, reasoning, and cybersecurity, though they are currently compute-intensive and expensive. Meta's Avocado models are delayed and may license Gemini technology, while Anthropic sees a surge in paid subscribers, nearly doubling this year. Separately, an open-source agent called AutoBe significantly improves function calling success rates for AI agents. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT New model capabilities from Anthropic and Meta signal continued progress, while AutoBe's function calling improvements could enhance AI agent reliability.
RANK_REASON The cluster discusses new model capabilities and performance benchmarks from multiple AI labs, along with an open-source agent improvement.