Back to home
TensorRT
2 articles tagged with this topic
NVIDIATensorRT
NVIDIA Cuts Open-Source Deployment to Two Commands — Convenience Is New Business
NVIDIA's TensorRT Model Connect deploys open-source LLMs in two commands. As GPUs become abundant, "making AI run" is itself a new business.
1d ago2 min read
NVIDIAAITune
NVIDIA Open-Sources AITune: Auto-Selects Fastest PyTorch Inference Backend
NVIDIA releases AITune, an open-source toolkit that benchmarks and auto-picks the optimal inference backend for PyTorch models.
Apr 122 min read