Skip to content

[Feature] GPT-OSS fp8 kv cache support #9782

@rainj-me

Description

@rainj-me

Checklist

Motivation

Currently GPT-OSS support triton and trtllm backend on SM100, but none of them support fp8 kv cahce.

Related resources

No response

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions