Guide

OpenRouter Token Usage Rankings Explained

How to read OpenRouter public model rankings and pricing data without confusing router volume for global model usage.

Updated 2026-05-12openrouter / pricing / open-models
Desk note

OpenRouter-style public rankings are useful because they are visible and model-specific. The risk is scope: a router ranking is not a claim about global usage unless the source explicitly says so.

What the data can show

Public router rankings can show which models are popular on that routing surface and how usage shifts alongside pricing, context, latency, and availability. That is useful directional signal.

  • Use it to compare model momentum on the router.
  • Use it to notice changes worth investigating.

ReceiptsOpenRouter rankingsOpenRouter apps

On this siteOur live leaderboard built on this data

What the data cannot prove

Router rankings are not global model usage unless the source explicitly makes that claim. Treat them as a public slice, not the whole market, and keep the scope visible anywhere the ranking is reused in a model, cost, or adoption argument.

  • Avoid phrases like total global token burn unless sourced.
  • Keep source URL and checked date next to derived claims.

Why pricing matters

A model can be popular because it is cheap enough, fast enough, available through a preferred API, or good enough for a specific workload. Ranking without pricing context can mislead.

  • Compare input and output price separately.
  • Look at context window and provider availability together.

Best use

Use the rankings to compare model cost and router momentum, then validate model choice against your own quality, latency, and acceptance metrics. Public rankings are a map, not your destination.

  • Let public rankings suggest tests.
  • Let your evals and traces decide production routes.
Weekly briefing

The term is moving faster than the definition.

Tokenmaxxing keeps shifting as new receipts land. The weekly briefing tracks who's burning what, and why it matters.

Written by the desk's AI, human-reviewed before send, real numbers only.

Source trail

Current feed records connected to this guide

Anthropic source artwork
newsA
news

Introducing Claude Fable 5.2 and Claude Mythos 5.2

Anthropic released Claude Fable 5.2 and Mythos 5.2 on Nov. 9, priced at $8 per million tokens in and $40 per million out, which it calls a lower price. An optional fast mode generates up to 2.5 times quicker at exactly double both rates.

tokenmaxxingpricingmodel-routing
Read note
Generated Tokenmaxxing editorial thumbnail for The AI boom has a new complication: falling prices
newsA
news

The AI boom has a new complication: falling prices

As of Oct. 22, two token-price gauges were sliding: Silicon Data's volume-weighted index fell under $1 per million tokens, 54% off its late-May high of $2.07, and Ramp's index of real business spend sat 48% below its February top of $1.22.

tokenmaxxingpricingai-spend
Read note
Generated Tokenmaxxing editorial thumbnail for The Hidden Cost of LLM Routing: Why Switching Models Costs More
long-formTD
long-form

The Hidden Cost of LLM Routing: Why Switching Models Costs More

Nhu Hoang replayed one 20-turn agent session under three routing policies on a vLLM server with prefix caching, then priced the token flows at Claude API rates. Switching models every turn saved 13%; the price sheet promised 40%.

tokenmaxxingmodel-routingmodel-router
Read note
Project layer

Tools that make the guide operational

#1Direct
Routing

LiteLLM

BerriAI/litellm

An OpenAI-compatible gateway and SDK for calling many model providers with budgets, logging, load balancing, guardrails, and cost tracking.

63.6K13.4KSource-available
gatewaycost-trackingrouting
#2Direct
Observability

Langfuse

langfuse/langfuse

Open-source LLM engineering platform for observability, traces, metrics, evals, prompt management, datasets, and playground workflows.

37.7K4.3KSource-available
tracesevalscosts
#5Direct
Evaluation

promptfoo

promptfoo/promptfoo

A CLI and CI workflow for testing prompts, agents, and RAG systems across models, with evals and red-team style checks.

27.5K2.8KMIT
prompt-evalscirag
Briefing

Fresh source notes each week.

New tokenmaxxing links, model-router signals, agent usage research, and AI cost notes.