zlm-v1-tool-decisions

zlm-v1-tool-decisions turns user intent into action, combining fast MCP tool selection with high-quality query and argument generation.

Built to help agents complete tasks efficiently, it understands the conversation, selects the right tools, and generates the inputs needed to use them. Decision-making and text generation work together in a single inference step, helping agents move from understanding a request to taking action.

When no tool is needed, it returns NO_TOOL, avoiding unnecessary calls and keeping workflows efficient. Purpose-built for agents that need to choose quickly and act precisely.