Managed Agents can now pin which region runs the compute

Claude Managed Agents take a new model.inference_geo parameter with two values. Set it to global and requests go wherever there’s free capacity, at standard pricing. Set it to us and the compute stays in region, billed at a 1.1x multiplier.
Per Anthropic’s documentation, the parameter locks the geography of requests to the model and is checked at every step of the agent’s run. It’s written up in the Claude Cookbook, dated July 23, 2026.
Up next
What 446 Reddit threads say about AI coding agents going rogue
A study of 446 Reddit threads on AI coding tools found 43.1% of security complaints were unauthorized file operations, led by agents deleting directories without permission (28.3%). One Claude Desktop user got messages from another user’s session.
Related stories
- Managed Agents get session budgets and an advisor model
- Claude Managed Agents get an AG-UI cookbook
- Claude Managed Agents get per-agent effort levels
- Claude apps switch to auto mode by default next week
- Claude Duet works while you’re away and asks before acting
- Claude Managed Agents now plans runs of up to 1,000 agents
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.

