DigestAI news desk

Cut through the AI noise.

Business & Funding2 min read

Fireworks reports serving more than 40t tokens per day

Interconnects’ dashboard tracks open model inference usage across the web, but the data is noisy and incomplete. The author notes that the best estimates come from a handful of startups—Baseten, Fireworks, and Together AI—whose public token‑volume disclosures show near‑exponential growth.

1 source

Key points

  • fireworks serving more than 40t tokens per day
  • baseten fireworks together ai used as bellwethers for open model inference
  • open model inference shows near‑exponential growth before GLM 5.2 step change

Fireworks announced at its first company conference that it is serving more than 40T tokens per day to customers’ customized models. The company said this figure grew over 2× from summer, suggesting a fall total well above 100T tokens. The author compares these numbers to closed‑model volumes from hyperscalers such as Amazon Bedrock and Azure’s model garden, noting that the latter do not break down usage by model category.

The notes also reference OpenRouter’s token volume share and Google’s earnings‑reported platform usage, which the author overlays with the open‑model data to illustrate the market’s size and trajectory. The article remains a research note, with the author acknowledging that the figures are approximations and that the dashboard will stay free for the foreseeable future.

The story so far

2 episodes →
  1. Fireworks reports serving more than 40t tokens per daythis story
Full story from Interconnects · by Nathan LambertOpen source ↗

How big is the open model economy today?

Interconnects · 4 November 2026

Loading the full article…

This text was published by Interconnects and written by Nathan Lambert. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Business & Funding

All →

Related stories