Skip to content

[feature]: support mtp layers coping to quantized folder #1477

@wenhuach21

Description

@wenhuach21

Problem Description

Current transformers discard this layer, however, this is required for vllm if mtp is enabled

Reproduction Steps

~

Environment Information

No response

Error Logs

Additional Context

No response

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions