OpenAIAnthropicGooglexAIPerplexity
← Back to all updates
Published ·Anthropic• Minor

Anthropic added an API option to disable parallel tool use by setting a specific boolean in tool_choice.

The read

Sequential tool execution prevents model errors in workflows where the second tool depends on the first output.

The frame

This update provides granular control for developers building complex agentic loops that require strict logic over speed.

Primary source
Anthropic
Keep reading
See what Nova3 builds
Related updates

This is the world we build in. It moves this fast. We move with it.

Bring us your project