May 31, 2026· ~9 min readTwo-Node DGX Spark Cluster: Running DeepSeek V4 Flash at 20 TPS Two NVIDIA DGX Sparks. One 685B-parameter MoE model. A 3× throughput improvement without a single crash. Here's how. ## The Setup: Two GB10 Workstat
May 28, 2026· ~6 min readTaming OOM on DGX Spark: Debugging Unified Memory Pressure in a 2-Node vLLM Cluster Your cluster loads the model. Weights land in GPU memory. Then — crash. `NV_ERR_NO_MEMORY`. The OOM killer takes down `VLLM::Worker_TP`. You restart.