OpenRouter's token usage plateaus as the token-maxxing era ends
Token-maxxing is dead, OpenRouter's token usage plateaus
OpenRouter's live rankings show weekly token usage flattening across its leaderboard, with DeepSeek V4.1 Flash still on top at 32.8T tokens but down 20%, while GLM 5.3 Flash holds flat at 12.2T. The data tracks tokens processed through the OpenRouter API across millions of developers, revealing where real-world LLM traffic is actually going — and where growth has stalled.
Models differ in verbosity and tokenization, so a higher token total shows how much a model is used, not which model is best for a task.