Google Deepmind adds background execution and MCP support to Gemini API managed agents
Google DeepMind is enhancing its Gemini API Managed Agents with background execution, MCP server connections, custom functions, and credential refresh capabilities.
- Agents in the Gemini API can now run asynchronously in the background, enabling long-running tasks without blocking.
- Direct MCP server connections allow agents to interact with external tools and services more seamlessly.
- Custom functions can now be used alongside sandboxed tools, increasing flexibility in agent operations.
- Credential refresh support ensures agents maintain state and access without interruptions.
Google DeepMind has introduced four new features for its Managed Agents in the Gemini API, significantly expanding their functionality. Agents can now run asynchronously in the background, allowing for long-running tasks without blocking the main process. This is particularly useful for workflows that require continuous monitoring or processing over extended periods.
The update also includes direct integration with remote MCP (Model Context Protocol) servers, enabling agents to interact with external tools and services more seamlessly. Additionally, developers can now combine custom functions with sandboxed tools, providing greater flexibility in how agents operate. Finally, the credential refresh feature ensures agents maintain their state and access without interruptions, even when credentials expire or need updating.
These enhancements reflect Google DeepMind's push to make its AI agents more robust and versatile for enterprise and developer use cases. The background execution and MCP support are especially notable, as they address common pain points in agent-based systems, such as state management and external tool integration.
Source: Google Deepmind adds background execution and MCP support to Gemini API managed agents. Read the full piece at the source.
Enables more robust, long-running agent workflows with seamless external tool integration and state management.
Improves operational efficiency by allowing agents to handle background tasks and external integrations without manual intervention.
Enhances the capabilities of AI agents in practical applications, making them more reliable for real-world use.
- MCP (Model Context Protocol)
- A protocol for enabling AI agents to interact with external tools and services.
- Managed Agents
- Pre-built AI agents in the Gemini API designed for specific tasks or workflows.
AI ToolsFlint: A visualization language for the AI era
AI ToolsManage AI applications on Mac with Jamf’s AI Governance and Amazon Bedrock
AI ToolsNVIDIA Nemotron Achieves Benchmark-Leading Performance With LangChain Deep Agents Harness
AI ToolsA New Developer Platform for Agent-Human Collaboration
AI helps doctors estimate your heart’s ‘age’ - KKTV
SecuritySecuring Amazon Bedrock AgentCore Runtime with AWS WAF
AWS details two secure patterns for routing Amazon Bedrock AgentCore Runtime traffic through AWS WAF and VPC Interface Endpoints, eliminating direct access risks.
Mistral launches first robotics model in physical AI push - Reuters
Mistral has launched its first robotics model, marking a significant step in its physical AI efforts. This move is part of the company's push into the robotics sector.
China to Let AI Firms Buy Nvidia H200 Chips, Information Says - Bloomberg
China will allow local AI companies to purchase Nvidia H200 chips, reversing prior restrictions on high-end AI hardware.
China plans to let top AI firms buy limited amount of Nvidia H200 chips, the Information reports - Reuters
China will allow select AI companies to buy a restricted quantity of Nvidia H200 chips, according to a report by The Information.
China Plans to Let Top AI Firms Buy Limited Amount of Nvidia H200 Chips - The Information
China will allow select AI companies to buy a restricted quantity of Nvidia H200 chips, according to a report from The Information.
AI helps doctors estimate your heart’s ‘age’ - KOLN | Nebraska Local News, Weather, Sports | Lincoln, NE
A new AI system helps doctors estimate a patient's heart age, potentially improving cardiovascular health outcomes. The system uses machine learning algorithms to analyze medical data.