Skip to content

Integer by zero crash when trying FP8 quantization inference #4817

Description

@kar-dim

NOTE: I use TensorRT-RTX, so naturally I submitted it on TensorRT-RTX repo, but I think it is more relevant here because it seems to be TRT specific and TRT-RTX is probably not to blame. All the information is there, I can "migrate" it here and close it from there if you want.

NVIDIA/TensorRT-RTX#35

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions