skills.rest
drdave-flexnetos
drdave-flexnetosCOMMUNITY
Explore all skills by this author →
GitHub

multi-model

Dispatch ARIA workflows across multiple LLM providers with model routing files.

Updated Jan 12, 2026
Search Tags:#llm routing#openrouter#ollama#claude-code-router#multi-model orchestration#openai-compatible api#hybrid agent dispatch
One-click install
npx skills add https://github.com/drdave-flexnetos/ripple-env --skill multi-model-drdave-flexnetos
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: multi-model
Source: https://github.com/drdave-flexnetos/ripple-env/tree/main/.claude/skills/multi-model
Command: npx skills add https://github.com/drdave-flexnetos/ripple-env --skill multi-model-drdave-flexnetos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Multi-model orchestration solves the problem of getting reliable outputs while managing cost, latency, and capability differences across LLM providers.

Core Features & Use Cases

  • Provider routing and dispatch: Choose between multiple models (e.g., Claude, OpenAI, Ollama/local, vLLM, LocalAI, Moonshot Kimi K2) based on task type and fallback rules.
  • Hybrid orchestration (recommended): Use Claude as the main orchestrator while delegating subagent/background work to Kimi K2 via claude-code-router.
  • Parallel and consensus workflows: Run multiple model perspectives in parallel and optionally combine results using agreement thresholds for critical decisions.

Use Case Example: For a codebase security review, route quick scanning to a fast model, route deep reasoning to a stronger model, and use local models as offline fallbacks while ensuring tool-using “thinking” models handle the trickiest sub-tasks.

Quick Start

Configure your model registry and routing in .claude/config/models.json, then activate hybrid routing with ccr activate and run claude to let the router dispatch subtasks to the selected providers.

Frequently Asked Questions about multi-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How does multi-model LLM routing improve response quality under cost and latency constraints?▼

Multi-model LLM routing improves response quality by dispatching a single workflow across multiple providers, selecting models based on task type and fallback rules to balance cost, latency, and capability. It enables parallel multi-perspective querying and consensus-based decision making across local and cloud backends.

How do I configure claude-code-router for hybrid orchestration with Kimi K2?▼

To configure hybrid orchestration with claude-code-router, set up your model registry and routing in .claude/config/models.json, then activate hybrid routing with the 'ccr activate' command and run 'claude' to let the router dispatch subtasks to selected providers like Kimi K2.

Can I use local Ollama models as offline fallbacks alongside cloud providers?▼

Yes, you can use local Ollama models as offline fallbacks alongside cloud providers. The routing configuration allows you to define fallback rules, ensuring that if a cloud provider is unavailable, tasks are automatically dispatched to local backends like Ollama, vLLM, or LocalAI.

What is the best way to run parallel multi-perspective querying across different LLM providers?▼

The best way to run parallel multi-perspective querying is by configuring a multi-model orchestration workflow that queries multiple LLM providers simultaneously and optionally combines their results using agreement thresholds for critical, consensus-based decisions.

Does multi-model orchestration support OpenAI-compatible APIs for custom backends?▼

Yes, multi-model orchestration supports OpenAI-compatible APIs. It enforces provider-specific API base URLs, authentication, and model selections through the routing configuration, allowing custom backends like vLLM and LocalAI to integrate seamlessly.

When should I use a hybrid Claude and Kimi K2 setup instead of a single LLM provider?▼

You should use a hybrid Claude and Kimi K2 setup when you need to optimize complex workflows, such as codebase security reviews, by using Claude as the main orchestrator for deep reasoning while delegating quick scanning and background subagent work to a faster model like Kimi K2.

Related Skills

Explore More Software Engineering Skills→
rauell1rauell1
3

ai-gateway

Configure Vercel AI Gateway routing, failover, and cost controls with OIDC authentication.

Community
Advanced
dralgorhythmdralgorhythm
111

reaching-consensus

Facilitate group decision-making using Fist of Five and Gradients of Agreement.

Community
Intermediate
FlexNetOSFlexNetOS

multi-model

Routes tasks across Claude, Kimi K2, OpenAI, and local models via configurable hybrid or subagent workflows.

Community
Advanced
NovaAI-innovationNovaAI-innovation

model-discovery

Fetch current model names from AI providers and classify them into performance tiers.

Community
Intermediate