Who serves the tokens?
Tokens are the AI economy's unit of account — every chat, coding agent and pipeline consumes them. This stacks estimated LLM inference volume by model family since 2019, the way forms of rule stacks humanity by regime: hover for a breakdown, and switch between relative share and absolute tokens per day.
Editorial, order-of-magnitude estimates — no official industry-wide token statistics exist. Anchors include vendor disclosures (e.g. Google's tokens-processed-per-month figures, Microsoft earnings mentions) and public router rankings; the split between families is inferred and approximate, and “tokens served” conflates consumer, API and internal workloads. Hover for a breakdown; switch between share and absolute volume above.
How to read it
Three stories sit in the stack. First, the explosion: from research-scale volumes in 2019 to tens of trillions of tokens a day by 2025 — the fastest adoption curve of any computing platform. Second, the erosion of the OpenAI monopoly: near-total dominance after ChatGPT's launch gives way to a contested oligopoly as Gemini rides Google's distribution, Claude takes a deep position in code and enterprise work, and DeepSeek and Qwen make frontier-adjacent capability nearly free. Third, the open-weights floor: Llama, Mistral and their descendants guarantee that some share of the world's tokens can never be enclosed — the “non-state societies” band of this chart.
Tokens are not revenue
This chart will mislead you if you read it as a market-share-of-money chart. By mid-2025, Menlo Ventures' enterprise survey had Anthropic ahead of OpenAI in enterprise LLM API spend (roughly 32% vs 25%, and about 42% of code-generation workloads) — while this chart still shows OpenAI and Google far ahead in tokens. Both can be true at once, because the token market is a barbell. Anthropic sells comparatively few, expensive tokens into paid API work — coding agents, enterprise pipelines — where revenue per token is highest. OpenAI's and Google's volumes are dominated by consumer and free-tier usage: hundreds of millions of chat users and AI-in-search answers that generate enormous token counts and little or no marginal revenue. Divide each family's revenue by its volume and the implied price per token differs by an order of magnitude. So: for where the money is, Anthropic's enterprise position is the story; for where the compute goes, consumer-scale distribution still rules this chart. The full argument is in The price of a token.