multi-model

Multi-model routing distributes tasks across Claude, Kimi K2, OpenAI, and local models to optimize cost and latency. You configure provider settings and router templates to orchestrate hybrid workflows and subagent tasks efficiently.

Configure hybrid routing by defining provider settings in the .claude/config directory. This config-driven approach manages model registries and router templates to dispatch tasks across Claude, Kimi K2, OpenAI, and local models.

Multi-model orchestration coordinates task distribution across multiple LLM providers to optimize capability selection, cost, and latency. You need it for hybrid routing scenarios or subagent workflows requiring flexible decision-making across different models.

Yes, mode-based orchestration allows switching between full replacement, hybrid, or subagent-only workflows. This flexibility lets you tailor the LLM orchestration strategy to match specific task requirements and provider capabilities.

The best way to manage model registries is using config-driven routing. By maintaining simple configuration files, you establish router settings that control how tasks are distributed across registered LLM providers.

Yes, the multi-model orchestration skill operates without external dependencies. It relies entirely on internal configuration files and router templates to manage the dispatch logic across your connected LLM providers.