NVIDIA has made its model transfer-learning and optimization toolkit "TAO 7" (TAO Toolkit 7.0/7.0.1) generally available. Shifting decisively from its CLI-centric design to an "agent-based" approach, the toolkit now lets developers fine-tune computer vision (CV) and vision language models (VLM) and tune hyperparameters simply by instructing a coding agent in natural language.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.