How to do inference for a FP16 model with non-fixed input shape in C++?

The model was trained with Python. I have looked into different ways but hit the wall here or there. I summarize as below and please correct me if I am wrong

+-----------------+-----+-------+-----------------+
|                 | C++ | FP 16 | non-fixed shape |
+-----------------+-----+-------+-----------------+
| MXNet C++ API   | ✓   | ?     | ✓               |
| TVM             | ✓   | ✓     | X               |
| MXNet-TensorRT  | X   | ✓     | ✓               |
+-----------------+-----+-------+-----------------+

So what’s the better practice for my case? Is the proposed fix in #14159 the only way to go?