Migrating Python Code from TensorRT 10.x to 11.x

This page describes how to update Python code when you migrate from TensorRT 10.x to 11.x: paired examples for strong typing, explicit quantization, plugin migration, and updated runtime APIs, followed by lists of Python APIs added and removed in 11.x.

Note

The Python API is not supported on QNX.

Warning

Engine files are executable artifacts that contain compiled CUDA tactics. The deserialize_cuda_engine examples in Migrating from Weak Typing to Strong Typing assume the engine file was built by you on a trusted host or received over a trusted, authenticated channel. Never deserialize an engine file from an untrusted source. Refer to Engine Deserialization and the Trust Boundary for the full trust-boundary model.

For conceptual background on strong typing, plugin V3, and other 11.x breaking changes, see Migrating from TensorRT 10.x to 11.x.

• Migrating from Weak Typing to Strong Typing — paired before/after Python examples

• Removed Python APIs and Replacements — removed Python APIs and replacements

• Migrating from Weak Typing to Strong Typing
  • Before (TensorRT 10.x)
  • After (TensorRT 11.x)
  • Summary of Changes
• Migrating INT8 Calibration to Explicit Quantization
  • Before (TensorRT 10.x)
  • After (TensorRT 11.x)
  • Summary of Changes
• Migrating Plugins from V2 to V3
  • Before (TensorRT 10.x)
  • After (TensorRT 11.x)
  • Summary of Changes
  • Known Issues When Migrating Plugins
• Migrating Weight Streaming APIs
  • Before (TensorRT 10.x)
  • After (TensorRT 11.x)
  • Summary of Changes
• Migrating Memory Management APIs
  • Before (TensorRT 10.x)
  • After (TensorRT 11.x)
  • Summary of Changes
• Removed Python APIs and Replacements