Smart Model Switching

Auto-route tasks to the cheapest Claude model that works correctly. Three-tier progression: Haiku → Sonnet → Opus. Classify before responding. HAIKU (default): factual Q&A, greetin…

millibus

@millibus

What This Skill Does

Auto-routes user tasks to the cheapest Claude model (Haiku, Sonnet, or Opus) based on complexity, starting with Haiku and escalating only when the task requires deeper reasoning. Uses a three-tier classification system with clear rules for when to escalate.

Replaces manual model selection and wasteful use of expensive models by automatically matching task complexity to the appropriate tier, saving 50-90% on API costs.

When to Use It

  • Route simple factual Q&A or reminders to Haiku to save costs
  • Escalate code generation over 10 lines or multi-step analysis to Sonnet
  • Send architecture decisions or complex debugging to Opus
  • Automate cost-optimized task routing in multi-agent workflows
  • Handle a mix of simple and complex user requests without manual model switching
  • Reduce API spending on routine monitoring or heartbeat checks by using Haiku

Install

$ openclaw skills install @millibus/smart-model-switching

Smart Model Switching

Three-tier Claude routing: Haiku → Sonnet → Opus

Start with the cheapest model. Escalate only when needed. Save 50-90% on API costs.

The Golden Rule

If a human would need more than 30 seconds of focused thinking, escalate from Haiku to Sonnet. If the task involves architecture, complex tradeoffs, or deep reasoning, escalate to Opus.

Cost Reality

ModelInputOutputRelative Cost
Haiku$0.25/M$1.25/M1x (baseline)
Sonnet$3.00/M$15.00/M12x
Opus$15.00/M$75.00/M60x

Bottom line: Wrong model selection wastes money OR time. Haiku for simple, Sonnet for standard, Opus for complex.


💚 HAIKU — Default for Simple Tasks

Stay on Haiku for:

  • Factual Q&A — "what is X", "who is Y", "when did Z"
  • Quick lookups — definitions, unit conversions, short translations
  • Status checks — calendar, file reads, session monitoring
  • Heartbeats — periodic checks, HEARTBEAT_OK responses
  • Memory & reminders — "remember this", "remind me to..."
  • Casual conversation — greetings, small talk, acknowledgments
  • Simple file ops — read, list, basic writes
  • One-liner tasks — anything answerable in 1-2 sentences

NEVER do these on Haiku

  • ❌ Write code longer than 10 lines
  • ❌ Create comparison tables
  • ❌ Write more than 3 paragraphs
  • ❌ Do multi-step analysis
  • ❌ Write reports or proposals

💛 SONNET — Standard Work (The Workhorse)

Escalate to Sonnet for:

Code & Technical

  • Code generation — write functions, build features, scripts
  • Code review — PR reviews, quality checks
  • Debugging — standard bug investigation
  • Documentation — README, comments, user guides

Analysis & Planning

  • Analysis & evaluation — compare options, assess trade-offs
  • Planning — project plans, roadmaps, task breakdowns
  • Research synthesis — combining multiple sources
  • Multi-step reasoning — "first... then... finally"

Writing & Content

  • Long-form writing — reports, proposals, articles (>3 paragraphs)
  • Creative writing — blog posts, descriptions, copy
  • Summarization — long documents, transcripts
  • Structured output — tables, outlines, formatted docs

❤️ OPUS — Complex Reasoning Only

Escalate to Opus for:

Architecture & Design

  • System architecture decisions
  • Major codebase refactoring
  • Design pattern selection with tradeoffs
  • Database schema design

Deep Analysis

  • Complex debugging (multi-file, race conditions)
  • Security reviews
  • Performance optimization strategy
  • Root cause analysis of subtle bugs

Strategic & Creative

  • Strategic planning — business decisions, roadmaps
  • Nuanced judgment — ethics, ambiguity, competing values
  • Deep research — comprehensive multi-source analysis

🔄 Implementation

For Subagents

```javascript // Routine monitoring sessions_spawn(task="Check backup status", model="haiku")

// Standard code work
sessions_spawn(task="Build the REST API endpoint", model="sonnet")

// Architecture decisions sessions_spawn(task="Design the database schema for multi-tenancy", model="opus") ```

For Cron Jobs

```json { "payload": { "kind": "agentTurn", "model": "haiku" } } ``` Always use Haiku for cron unless the task genuinely needs reasoning.


📊 Quick Decision Tree

``` Is it a greeting, lookup, status check, or 1-2 sentence answer? YES → HAIKU NO ↓

Is it code, analysis, planning, writing, or multi-step? YES → SONNET
NO ↓

Is it architecture, deep reasoning, or critical decision? YES → OPUS NO → Default to SONNET, escalate if struggling ```


📋 Quick Reference Card

``` ┌─────────────────────────────────────────────────────────────┐ │ SMART MODEL SWITCHING │ │ Haiku → Sonnet → Opus │ ├─────────────────────────────────────────────────────────────┤ │ 💚 HAIKU (cheapest) │ │ • Greetings, status checks, quick lookups │ │ • Factual Q&A, definitions, reminders │ │ • Simple file ops, 1-2 sentence answers │ ├─────────────────────────────────────────────────────────────┤ │ 💛 SONNET (standard) │ │ • Code > 10 lines, debugging │ │ • Analysis, comparisons, planning │ │ • Reports, proposals, long writing │ ├─────────────────────────────────────────────────────────────┤ │ ❤️ OPUS (complex) │ │ • Architecture decisions │ │ • Complex debugging, multi-file refactoring │ │ • Strategic planning, deep research │ ├─────────────────────────────────────────────────────────────┤ │ 💡 RULE: If a human needs > 30 sec thinking → escalate │ │ 💰 COST: Haiku 1x → Sonnet 12x → Opus 60x │ └─────────────────────────────────────────────────────────────┘ ```


Built for Claude-only setups with Haiku, Sonnet, and Opus. Inspired by save-money skill, extended with three-tier progression.

Top skills in this category

Humanizer

@biostartechnology

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.

706126k

Elite Longterm Memory

@nextfrontierbuilds

Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.

23264k

Model Usage

@steipete

Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.

11339k

description: 将用户讲稿一键生成乔布斯风极简科技感竖屏HTML演示稿。当用户需要生成PPT、演示文稿、Slides、幻灯片,或要求科技风/极简风/乔布斯风格的演示时触发此技能。输出为单个可直接运行的HTML文件。

@wwlyzzyorg

将讲稿一键生成乔布斯风极简科技感竖屏HTML演示稿

8533k

Proactive Agent Lite

@bestrocky

Transform AI agents from task-followers into proactive partners with memory architecture, reverse prompting, and self-healing patterns. Lightweight version f...

6240k