Skip to content
/
OpenRouter
© 2026 OpenRouter, Inc

Product

  • Chat
  • Rankings
  • Apps
  • Models
  • Providers
  • Pricing
  • Enterprise
  • Labs

Company

  • About
  • Announcements
  • CareersHiring
  • Privacy
  • Terms of Service
  • Support
  • State of AI
  • Works With OR
  • Data

Developer

  • Documentation
  • API Reference
  • SDK
  • Status

Connect

  • Discord
  • GitHub
  • LinkedIn
  • X
  • YouTube

Pareto Code Router

openrouter/pareto-code

Released Apr 21, 2026200,000 contextPricing varied

The Pareto Router is a way to have OpenRouter always pick a strong coding model for your needs without committing to a specific one. You express a single min_coding_score preference between 0 and 1, and the router routes your request to a coding model that meets that bar.

The Pareto Router is tuned for coding use cases. Under the hood it keeps a curated shortlist of strong coding models currently available on OpenRouter, ranked by Artificial Analysis(opens in new tab) coding scores. The exact shortlist and selection logic evolve over time as new models land and benchmarks shift.

Overview
Router
Activity
API

How routing works

How the Pareto Router selects coding models for your request.

The Pareto Router maintains a tiered shortlist of strong coding models, ranked by Artificial Analysis coding percentiles. Set min_coding_score between 0 and 1 on the pareto-router plugin to control how strong a coder you need; higher scores select stronger (and typically more expensive) models. If you omit min_coding_score, the router defaults to the High tier.

Today scores map into three bands — Low, Medium, and High. More granular bands will be added over time, so the precise score you pick now lets you opt into finer control as the shortlist grows.

Selecting Nitro from the variant dropdown ranks the models in your tier by measured throughput and routes each request to the fastest one, so you trade some model variety for lower latency.

Read the Pareto Router docs for the full selection logic, fallback behavior, and how to customize routing.

Models we route to

  • High tiermin_coding_score ≥ 0.66
    Favicon for openai
    gpt-5.5
    Favicon for google
    gemini-3.1-pro-preview
    Favicon for anthropic
    claude-opus-4.7
    Favicon for deepseek
    deepseek-v4-pro
  • Medium tier0.33 ≤ min_coding_score < 0.66
    Favicon for openai
    gpt-5.4-mini
    Favicon for anthropic
    claude-sonnet-4.6
    Favicon for moonshotai
    kimi-k2.6
    Favicon for x-ai
    grok-4.3
  • Low tiermin_coding_score < 0.33
    Favicon for xiaomi
    mimo-v2.5-pro
    Favicon for qwen
    qwen3.6-max-preview
    Favicon for z-ai
    glm-5.1
    Favicon for deepseek
    deepseek-v4-flash
    Favicon for anthropic
    claude-haiku-4.5