41T tokens served in a day: startup claims first quadrillion tokens/month for open-weight models
ycombinator · x · 2026-11-18
thdxr (amplified by the Y Combinator account) reports his inference business served 41 trillion tokens in a single day, and claims they'll be the first to process a quadrillion tokens per month serving open-weight models. The figure offers a rare concrete datapoint on how much real production traffic open-weight model inference now handles, and signals that open-weight inference is scaling into a token-throughput business at massive volume.
More from Infra
- 103M Gaussians run natively and smoothly on iPhone 15 Pro in first 3D scan tests — janusch_patas · 2026-11-19
- Weeks after Jensen Huang's reported firing, TSMC reportedly demands cash up front for orders — RachelVT42 · 2026-11-19
- 10 wild data center concepts: orbit, seabed, oil rigs, wind turbines and even swimming pools — clemnt · 2026-11-19
- Intel CEO Lip-Bu Tan Says a CEO Came to His Office Demanding Huge EMIB-T Supply — BenBajarin · 2026-11-19
- Analyst: US-made advanced silicon and memory is inevitable as Intel, Micron pull domestic demand — BenBajarin · 2026-11-19
- Intel CEO says Microsoft will use EMIB-T advanced packaging — BenBajarin · 2026-11-19