During the week of November 9‑15 several AI models and tools were released. Anthropic presented a model that can coordinate thousands of agents for scientific tasks, Google moved its Gemini system from chat‑only prototypes toward real‑world applications, and Meta launched a compact model designed for agent‑based scenarios. Benchmark developers also announced efforts to evaluate AI performance on more realistic tasks.
The summary gathers the main releases, research papers, tools and a glimpse of industry practices that appeared in that period.