
DSH Dynamic Router Intelligent Routing & Cost Balancer
An adaptive task-complexity model router for DeepSeek Harness. Automatically routes lightweight tasks to fast small models while escalating deep logical reasoning to DeepSeek-R1, saving costs.
Model-Routing · Cost-Optimization · Adaptive-Router · Complexity-Gate
$ pnpm add @yjh051108/dsh-routing-suiteArchitecture
Core Features
Dual-Metric Complexity Scoring: Micro-second evaluation of semantic ambiguity and dependencies
Seamless Fallback Escalation: Auto-promotes turns to flagship models if small models fail validation
Fine-Grained Budget Caps: Set hard expenditure ceilings per session or billing cycle easily
Live Traffic Analytics Dashboard: Real-time visual tracking of token savings and tier distributions
Core Workflow
Complexity Scoring
Calculates complexity scores based on prompt tokens, ambiguity, and tool schemas.
Rule-Based Matching
Maps evaluated scores to configured model tiers (Tier-1, Tier-2, or Reasoning-Tier).
Payload Forwarding
Adjusts temperature and token parameters per target provider before dispatch.
Quality Gate Fallback
Transparently escalates to flagship models if small model outputs fail verification.
Configuration Parameters Reference (YAML / JSON)
| Parameter | Type | Default | Description |
|---|---|---|---|
| budgetMode | string | balanced | Routing strategy prioritization |
| complexityThreshold | number | 0.65 | Threshold to escalate to DeepSeek-R1 |
| lightweightModel | string | deepseek-chat-fast | Target lightweight model endpoint |
Recommended Plugins
dsh-our-free-model Keyless Free Model Gateway Plugin
dsh-our-free-model is a keyless model provider plugin for DeepSeek Harness. Requiring no account, sign-up, or API key, it provides instant access to frontier models like DeepSeek V4.1 and Kimi K3. Featuring real token budget thinking control, egress network probing, and a local OpenAI-compatible forward port, it keeps all telemetry safely on your machine.
OpenRouter Live Multi-Model Gateway & Failover Router
A unified multi-model routing gateway connecting DeepSeek Harness with hundreds of global LLM providers via OpenRouter. Features automated failover, latency routing, and unified billing.
AGY Dedicated Inference Provider & Connection Pool
Official AGY Cloud dedicated inference driver for DeepSeek Harness. Features HTTP/2 multiplexed persistent connections, binary streaming protocols, and ultra-low latency cluster routing.