DeepSeek serves 150T tokens/day on one cluster, more than OpenRouter, Fireworks, Together and BaseTen combined
natolambert · x · 2026-10-24
Nathan Lambert reconstructed open-model inference volumes from company reports: DeepSeek claims to serve 150T tokens per day on a single cluster, exceeding open-model inference on OpenRouter, Fireworks, Together, and BaseTen combined. The accompanying chart shows Fireworks and Together reaching roughly 40T and 30T tokens/day respectively by 2026, with OpenRouter's open-model share climbing from 20B to 10T+ tokens/day. Commenters note this is just "one cluster" — DeepSeek's true total scale is unknown.
More from Infra
- Tencent's Hy-Nomas world model needs at least 24GB VRAM, dev complains — cocktailpeanut · 2026-10-24
- Humansand Speeds Up NVFP4 Decode Kernel by Up to 10% While Improving Accuracy — noahdgoodman · 2026-10-24
- Recursive's 40B Flint Hits ~200 tok/s on M3/M5 Ultra via MLX-VLM with Fused Metal Kernels — timshi_ai · 2026-10-24
- LiteLLM breach lesson: AI gateway logs are a secrets store — alvelda · 2026-10-24
- US electricity prices up 34% at investor-owned utilities since 2020 vs 19% public — randal_olson · 2026-10-24
- ExecuTorch now builds on Windows Snapdragon X Elite, demoing on-device Llama 3.2 1B — carrycooldude · 2026-10-24