Updated GPT-4 Turbo now available
We have enabled the latest version of GPT-4 Turbo (0125 / Jan 25th) in all workspaces. This new build features improved reasoning, less "laziness" and a more recent knowledge cut off. We recommend testing your prompts before switching models. You can set the LLM step in your workflow to either the latest available version of GPT-4 Turbo, or a specific version.
Note : In early testing, this build is still not surpassing GPT-4 32k in extremely complex or longer-form tasks but is the best choice in the majority of complex use cases.