Anthropic’s upgraded Console targets more collaboration among developers

Anthropic has upgraded its Console — an interface that lets enterprises create API keys, add more users to a team, set up billing, and experiment with Claude on the Workbench, in an effort to promote more collaboration.

“Developers can now share prompts to collaborate with teammates directly in the Anthropic Console,” the company wrote in a blog post, adding that previously they had to resort to copying and pasting prompts between documents or chat applications, leading to version control issues and knowledge silos.

The capability is expected to help enterprises use AI more effectively in their applications and other use cases, according to Mansi Gupta, practice director at Everest Group.

“Teams can now collaborate more effectively, share best practices, and keep a running library of prompts, so nothing gets lost over time. That means no more knowledge silos, and business teams can easily find and use the best prompts,” Gupta explained.

Another added feature to the upgraded Console includes optimizing prompts for extended thinking.

Unlike the instant responses from Claude, extended thinking shows the user all the steps the model has taken to reach an output or response.

“While prompting generally works the same with extended thinking enabled, we’ve made it easy to optimize prompts to make the most of this new capability,” the company wrote, adding that users would need to specify which prompts should work with extended thinking.

In addition, Anthropic is also offering a budgeting capability for extended thinking inside the Console, which can be used to set a limit to the maximum amount of “thinking” tokens being generated.

The budgeting capability, according to Gupta, will allow enterprises to become better at problem-solving without the risk of overspending. “With this feature, enterprises gain control over when they want their models to think and by how much,” Gupta said.

However, Amalgam Insights chief analyst Hyoun Park pointed out that the thinking or reasoning capability is not unique to Anthropic and can also be found in offerings from rivals, such as OpenAI.

Other updated features include automatically generating prompts, evaluating model responses, and improving existing prompts with Claude that were written for other AI models or were manually keyed-in.

In order to automatically generate prompts, users need to put in a natural language input in order to allow Claude to come up with “reliable and precise” prompts, the company wrote. Similarly, enterprises can also evaluate model responses on a set of prompts by running test suites inside the Console, which will allow them to grade response quality, it added.

More Anthropic news:

Anthropic unveils new framework to block harmful content from AI models

Anthropic introduces the Model Context Protocol

Anthropic adds prompt caching to Claude, cutting costs for developers