on Ampere based GPUs.TensorFloat-32, or TF32 for short, is a math mode
        for NVIDIA Ampere based GPUs which causes certain float32 ops, such as
        matrix multiplications and convolutions, to run much faster on Ampere
        GPUs but with reduced precision. This reduced precision has not been
        found to effect convergence quality of deep learning models in practice.

                          type: object
                        delay:
                          description: Delay requests before forwarding, emulating
                            various failures such as network issues, overloaded upstream
                            service, etc.
                          oneOf:
                          - not:
                              anyOf:

                          type: object
                        delay:
                          description: Delay requests before forwarding, emulating
                            various failures such as network issues, overloaded upstream
                            service, etc.
                          oneOf:
                          - not:
                              anyOf: