https://minerka.com/library/infrastructure/ Chips, clusters, inference servers, and the cost of running a model. Filed when a spec or a price comes from the company that sells it. Mon, 02 Nov 2026 11:29:44 +0000 https://minerka.com/stories/cuda-13-4-gpu-fraction/ https://minerka.com/stories/cuda-13-4-gpu-fraction/ On October 27, 2026, NVIDIA said CUDA 13.4 enforces a per-workload GPU memory boundary. A job can ask for an exact size, and an allocation past that size fails. KAI Scheduler v0.18.0 can write that limit from Kubernetes. Time-slicing keeps each job in its own error domain. The parallel mode does not. Start here (ranks, prices, unknown): https://minerka.com/learn/ October 27, 2026 https://minerka.com/stories/deepinfer/ https://minerka.com/stories/deepinfer/ DeepSeek's deployment note says it is open-sourcing DeepInfer, the inference framework behind V4.1 Flash. The repository is Apache 2.0. DeepSeek says one cluster serves 155 trillion tokens a day, and that the KV cache and the load balancer are not in the open code. Start here (ranks, prices, unknown): https://minerka.com/learn/ October 20, 2026