Claude AI got new model selector, Context Connectors in development

Recent reports from TestingCatalog and Tibor provide insights into the ongoing development of Anthropic's Claude AI, specifically focusing on the latest updates and upcoming features.

TestingCatalog highlighted that the older Claude 3.5 Sonnet model is now accessible via an updated model selector. This update allows users to switch between different Claude models more easily, offering flexibility depending on their needs. The newer Claude 3.5 Sonnet model, described as Anthropic's most intelligent to date, is also available, alongside other models like Claude 3 Opus (for creative tasks) and Claude 3 Haiku (a faster model for daily tasks). However, the newest version of Claude 3.5 Haiku is not yet available on Claude.

Another significant development is the introduction of the Model Context Protocol (MCP), which was mentioned in @btibor91's report. MCP aims to provide a standardized way for applications to supply context to large language models (LLMs) like Claude. This protocol separates context provision from LLM interaction, potentially allowing more efficient integration with external systems. The MCP Python SDK supports both client and server capabilities, making it easier for developers to build applications that can interact with LLMs using this protocol.

These updates reflect Anthropic’s broader strategy of enhancing Claude’s versatility across different use cases, from creative writing to complex data analysis. The introduction of MCP suggests a focus on improving how external data and tools integrate with LLMs, potentially expanding Claude’s utility in professional environments.