V0.11.0 Release Plan

[cp5555]

Release Manager

@cp5555

Endgame

• [ ] Code freeze: TBD
• [ ] Bug Bash date: TBD
• [ ] Release date: TBD

Main Features

SuperBench Improvement

1. • [x] Add CUDA 12.4 dockerfile (#619)

Micro-benchmark Improvement

1. • [x] Add hipblasLt tuning to dist-inference cpp implementation (#616)
2. • [x] Upgrade mlc to v3.11 (#620)
3. • [ ] Support cuDNN Backend API in cudnn-function.

Model Benchmark Improvement

1. Support VGG, LSTM, and GPT-2 small in TensorRT Inference Backend
2. Support VGG, LSTM, and GPT-2 small in ORT Inference Backend
3. Support more TensorRT parameters (Related to #366)

Result Analysis