Upgrading onnxruntime-gpu Alone Speeds SAM 3 Inference 3.8x

A developer reported a free 3.8x inference speedup on a SAM 3-based malaria detection pipeline simply by upgrading onnxruntime-gpu to 1.31 with CUDA 13, with identical outputs and unchanged model and hardware.

2026-11-03 ~ 2026-11-03 · 2 related posts

• Just upgrading onnxruntime-gpu 1.23→1.31 with CUDA 13 made SAM 3 inference 3.8x faster — MaziyarPanahi · 2026-11-03
• Upgrading onnxruntime-gpu 1.23→1.31 made SAM 3 inference 3.8x faster on one L4 — MaziyarPanahi · 2026-11-03