Extended Data Table 1 Number of parameters for each model according to Azure

From: Comparative benchmarking of the DeepSeek large language model on medical tasks and clinical reasoning