Skip to content

browse LLM provider APIs and filter/sort by throughput, cost, quant, context and more. powered by openrouter

Notifications You must be signed in to change notification settings

sculptdotfun/modelgrep

Repository files navigation

OpenRouter Explorer

A web-based tool for exploring and comparing OpenRouter models by throughput, latency, price, context length, and more.

Quick Start

cd openrouter-explorer
pip install -r requirements.txt
python server.py

Then open http://localhost:8765 in your browser.

Features

  • Filter by: throughput, latency, price, context length, modality, provider
  • Sort by: any metric, ascending or descending
  • Search: filter models by name or ID
  • Real-time data: fetches live metrics from OpenRouter API
  • Caching: 5-minute cache to avoid rate limits

API Endpoints

  • GET /api/models - List models with optional filters

    • q - Search query
    • min_throughput - Minimum throughput (tokens/sec)
    • max_latency - Maximum latency (ms)
    • max_price - Maximum input price ($/1M tokens)
    • min_context - Minimum context length
    • modality - Filter by modality (text, image, audio)
    • provider - Filter by provider name
    • sort - Sort field
    • desc - Sort descending (true/false)
    • limit - Results per page
    • offset - Pagination offset
  • GET /api/stats - Get aggregate statistics

  • GET /api/refresh - Force refresh data from API

Environment Variables

  • PORT - Server port (default: 8765)

About

browse LLM provider APIs and filter/sort by throughput, cost, quant, context and more. powered by openrouter

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •