The new proprietary large language model, named Thomson, was introduced by Thomson Reuters on Monday. The company reported spending $40 million on talent and computing resources to develop the model, which was built from an open-source foundation.
log in to read full article