dsh-model-orchestrator: A General-Purpose Model Orchestrator for DeepSeek Harness
dsh-model-orchestrator is a general-purpose model orchestration plugin for DeepSeek Harness, designed to address the challenges of manual routing decisions in multi-model environments. By automatically discovering models, analyzing their capabilities, and performing intelligent routing, it enables efficient model scheduling. Key features include model discovery, capability analysis, task matching, and orchestration based on a “captain” role. After installation, it works automatically without complex configuration, supports intervention via commands or interfaces, and focuses on leveraging existing host capabilities rather than task management. It is suitable for the automatic decomposition and execution of complex tasks.
Read Moredsh-agentic-router: A Learning-Based Intelligent Routing Plugin for DeepSeek Harness
dsh-agentic-router is an intelligent routing plugin for DeepSeek Harness (DSH), designed to automatically select the optimal model tier based on task complexity. It employs a four-expert architecture (Rule-based, UCB, LinUCB, and k-NN) with an EXP3 meta-selector, and continuously optimizes its routing policy in real time through a data flywheel mechanism driven by quality signals such as failures, cost, and latency. The new version supports actual model switching and step-level routing while providing various control tools. It defaults to shadow mode and can be switched to active mode via configuration or commands. All decision and reward data are persisted to disk, enabling replay to restore the learning state, and human feedback correction is supported through the Web UI.
Read More