96 B200 GPUs. 12 nodes. A load balancer distributing requests across them.
📌 Подробнее: https://medium.com/google-cloud/1-million-tokens-per-second-qwen-3-5-27b-on-gke-with-b200-gpus-161da5c1b592
MemOps 🤨
Post #7395
1.65K
1 Million Tokens Per Second: Qwen 3.5 27B on GKE with B200 GPUs
📌 Подробнее: https://medium.com/google-cloud/1-million-tokens-per-second-qwen-3-5-27b-on-gke-with-b200-gpus-161da5c1b592
MemOps 🤨
📌 Подробнее: https://medium.com/google-cloud/1-million-tokens-per-second-qwen-3-5-27b-on-gke-with-b200-gpus-161da5c1b592
MemOps 🤨
