Darkbloom serves ~27B tokens a day on idle Macs worldwide, Gemma demand doubles
gajesh · x · 2026-11-21
gajesh announced Darkbloom hit an all-time-high in inference served — nearly 27B tokens in a day, all running on idle Mac machines around the world.
- @Spangler3000 led work improving the inference engine, reliability and model weights
- Darkbloom integrated OpenRouter's new routing algorithm into its routing system
- OpenRouter's dashboard shows Gemma 4 26B A4B leading at 15.7B tokens/day; Gemma demand grew 2x, signaling rising open-weight model usage
More from Infra
- Tenstorrent hits 5,300 tokens/sec/user decoding gpt-oss-120b on just 8 Galaxy units — DavidBennett__ · 2026-11-21
- KV Cache Rules Everything: agent context re-reads cost you a lot, providers almost nothing — AccBalanced · 2026-11-21
- Open-source router factors prompt cache into LLM switching decisions for coding agents — Broad_Abies9390 · 2026-11-21
- Inkling 2 Mini Support Lands in llama.cpp via PR #32694 — No_Algae1753 · 2026-11-21
- US chip crackdown on Southeast Asia hands Ambani and Adani a $200B AI infrastructure windfall — nordicinst · 2026-11-21
- Nvidia's $2.34T market cap now rivals Germany's entire stock market — HanchungLee · 2026-11-21