On August 27, NVIDIA disclosed that it had recorded a $400 million charge in the first half of the year due to excess H200 inventory.

The H200, NVIDIA’s flagship accelerator chip designed for high-performance computing (HPC) and large-scale AI model training, has previously enjoyed consistently strong market demand. The emergence of excess inventory and the resulting substantial charge indicate that the supply-and-demand dynamics of the high-end AI accelerator market are undergoing a period of adjustment.