Skip to content

Expose GenAI context and prefill lengths in winml build - #1463

Open
Qiong Wu (qiowu) (DingmaomaoBJTU) wants to merge 1 commit into
mainfrom
dingmaomaobjtu-expose-genai-lengths
Open

Qiong Wu (qiowu) (DingmaomaoBJTU) wants to merge 1 commit into
mainfrom
dingmaomaobjtu-expose-genai-lengths

Conversation

@DingmaomaoBJTU

Copy link
Copy Markdown
Collaborator

Summary

  • add positive-integer --max-cache-len and --prefill-seq-len options to winml build
  • forward optional overrides through the architecture-agnostic GenAI bundle dispatch while preserving recipe defaults when omitted
  • reject the bundle-only options on generic builds instead of silently ignoring them
  • update build and Qwen3 bundle documentation with the one-command workflow
  • add focused CLI coverage for defaults, forwarding, validation, and generic-build rejection

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant