Skip to content
AI IntelligenceSep 17, 2026AI Intelligence
Article

NVIDIA TensorRT Edge-LLM runs MLPerf benchmark 6.4x faster on Jetson AGX Thor

NVIDIA's TensorRT Edge-LLM has completed the MLPerf Edge Agentic Benchmark 6.4 times faster on the Jetson AGX Thor platform. This advancement supports the transition of AI agents from cloud data centers to edge devices like vehicles and robots.

Frontier EditorialSource: NVIDIA Generative AI
01

Source Brief

NVIDIA TensorRT Edge-LLM runs MLPerf benchmark 6.4x faster on Jetson AGX Thor: NVIDIA's TensorRT Edge-LLM has completed the MLPerf Edge Agentic Benchmark 6.4 times faster on the Jetson AGX Thor platform. This advancement supports the transition of AI agents from cloud data centers to edge devices like vehicles and robots.