XNews

NVIDIA Releases TensorRT Model Connect Public Preview for Two-Command Inference

NVIDIA Releases TensorRT Model Connect Public Preview for Two-Command Inference

NVIDIA released TensorRT Model Connect in Public Preview. A supported Hugging Face model can run end-to-end TensorRT inference in two commands, without intermediate ONNX export, and the resulting bundle works through native C++ APIs. The project was built with OpenAI Codex agents, with humans directing and reviewing model implementations, performance tuning, tests, integrations, and docs. It is open source.

Source posts