DeepSeek Processes 155T Tokens Daily on ~2,600 GB200s — Global Inference Scale May Vastly Exceed Estimates
teortaxesTex · x · 2026-10-29
TeortaxesTex shares his own usage: a "billion token" task (2B) on DeepSeek v4.1-flash cost just $6.08 at peak pricing, thanks to context caching saving 98%. With fewer than 500 users like him, that's already "a trillion" tokens.
The core math: DeepSeek reportedly serves 155T tokens/day on ≈2,600 GB200s — roughly 16 GPUs per trillion tokens per day. Extrapolating, he argues total token consumption across frontier models must be far larger than commonly assumed, since "a trillion is a small number."
Related event: Analysis: OpenRouter's token volume vastly overstates real AI compute needs(4 posts)→
More from Infra
- PGlite crosses 30 million weekly downloads as in-browser Postgres momentum builds — matei_zaharia · 2026-10-29
- Prime Inference scales from zero to 120B tokens per day on OpenRouter in one month — xeophon · 2026-10-29
- SJTU's TAO Cuts LLM Agent Serving Latency Up to 26.8% With Tool-Aware Joint Scheduling — AIOR-Research · 2026-10-29
- Nscale's unproven CEO chases $30 billion neocloud IPO amid AI turmoil — bloomberg · 2026-10-29
- Qwen3.8 Flash Next at IQ3_S hits 275 tok/s on a single RTX 5080 via Basalt fork — yehyakar · 2026-10-29
- Google Cloud CEO: Apple Scaling Foundation Model Pretraining on Latest TPUs — shayan_ · 2026-10-29