๋ณธ๋ฌธ์œผ๋กœ ๊ฑด๋„ˆ๋›ฐ๊ธฐ
TakoAPI
์Šคํ‚ฌ๋กœ ๋Œ์•„๊ฐ€๊ธฐ

model-manager

notestone ์ œ์ž‘

Productivity & Tasks
โ€”์•„์ง ํ‰๊ฐ€ ์—†์Œ
๋‚ด ํ‰๊ฐ€

์š”์•ฝ

**๐Ÿ’ฐ Optimize Your API Costs: Route Simple Tasks to Cheaper Models.**.

์„ค๋ช…

์ด ์Šคํ‚ฌ์ด ํ•˜๋Š” ์ผ

Routes AI tasks to cost-appropriate models based on task complexity. Fetches real-time pricing from OpenRouter and classifies tasks into simple, coding, creative, or complex categories. Integrates with a companion benchmarks skill for data-driven model selection.

Instead of manually tracking model pricing and capabilities across providers, it automates routing decisions and updates cost data in real time.

์‚ฌ์šฉ ์‹œ์ 

  • Routing translation jobs to cheaper models to cut API costs
  • Comparing current OpenRouter model prices before starting a project
  • Setting a budget cap with alerts for a team's monthly API spend
  • Generating a 30-day cost report to audit model usage
  • Automatically switching to a stable fallback when a cheap model fails repeatedly

์˜ˆ์‹œ ์›Œํฌํ”Œ๋กœ

์ž…๋ ฅ

User asks: plan 'translate this product description to French'

์—์ด์ „ํŠธ
  1. 1

    Fetches current model pricing from OpenRouter

  2. 2

    Classifies the task as simple content work

  3. 3

    Identifies GPT-4o-mini or Gemini Flash as cost-optimal choices

  4. 4

    Returns recommended model with estimated cost and savings percentage

์ถœ๋ ฅ

Routing recommendation: use GPT-4o-mini at $0.60/1M tokens (96% cheaper than premium models)

์š”๊ตฌ ์‚ฌํ•ญ

  • OpenRouter API key

์„ค์น˜

clawhub install notestone/model-manager

CLI ๋ช…๋ น์„ ์‹คํ–‰ํ•˜๊ฑฐ๋‚˜, ํ”„๋กฌํ”„ํŠธ๋ฅผ Claude Code, Cursor, Codex ๋“ฑ ์‚ฌ์šฉํ•˜๋Š” ์ฝ”๋”ฉ ์—์ด์ „ํŠธ์— ๋ถ™์—ฌ๋„ฃ์œผ์„ธ์š”.

๋งํฌ