Anthropic is about to break its silence: The Claude-maker is reportedly gearing up to release new versions of Sonnet and Opus that combine reasoning and tool use for the first time. That means they can propose an idea, then go check it themselves using external databases and apps. The same goes for coding: The models will test out what they’ve generated partway through and fix their mistakes before delivering a final output. We don’t have an exact timeline, but we expect the models to be out within the next few weeks.
You are viewing a single comment's thread from: