Midjourney, the independent research lab that provides amazing text-to-image generative AI, has selected Google Cloud as its infrastructure provider of choice for training its large-scale AI models.
As part of the partnership, Midjourney has employed Google Cloud's custom-developed AI accelerators, Tensor Processor Units (TPUs), to train its fourth generation AI model. In combination with Google Cloud's GPU VMs, which run on NVIDIA GPUs, Midjourney can render the generated images on the platform with breathtaking speed.
"As AI opens the door for innovation across industries, companies like Midjourney are choosing Google Cloud to take advantage of our open, flexible, and performant infrastructure, which runs with incredible scale and efficiency," said Amin Vahdat, VP/GM Systems and Services Infrastructure, Google Cloud. "We are excited to see the Midjourney community create impressive imagery thanks to our industry-leading offerings."
