Fireworks serves over 40T tokens daily, mostly customer-specialized post-trained models
natolambert · x · 2026-11-04
Nathan Lambert notes that inference provider Fireworks serves more than 40 trillion tokens per day of "specialized models," which he defines as any post-trained model built for a customer. The figure suggests custom post-trained models already account for a large share of real-world inference traffic.
More from Infra
- Big tech has contracted over 10 GW of nuclear power, but contracts aren't generation — ingliguori · 2026-11-04
- Fully Local AI Short Film Rendered on a Mac, Workflow Open-Sourced as Phosphene — Opening-Ad5541 · 2026-11-04
- Anthropic's 15-person chip team uses Claude to design its own AI chips — kimmonismus · 2026-11-04
- Anthropic's 15-person chip team uses Claude to design its own AI chips ahead of $1.5T IPO — kimmonismus · 2026-11-04
- AAOI signs $194M InP wafer supply deal with Yunnan Germanium at ~$1,077 per wafer — zephyr_z9 · 2026-11-04
- Google Cloud taps Tetrate to extend Gemini platform AI routing to multi-cloud and on-prem — IanAndrewsDC · 2026-11-04