Skip to content

feat: bind Qwen Image 2.1 INT8 Metal runtime - #1647

Open
q6y6y6 wants to merge 2 commits into
ModelTC:mainfrom
q6y6y6:qwen-int8-metal-runtime
Open

q6y6y6 wants to merge 2 commits into
ModelTC:mainfrom
q6y6y6:qwen-int8-metal-runtime

Conversation

@q6y6y6

@q6y6y6 q6y6y6 commented Oct 22, 2026

Copy link
Copy Markdown
Contributor

Summary

Add model-bound Qwen Image 2.1 Turbo INT8 Metal runtime support.

Changes

  • Add the int8-metal MPS configuration for Qwen Image 2.1 Turbo.
  • Add runtime validation for the model-bound Apple Metal dependency manifest.
  • Validate model repository and revision metadata.
  • Validate required K/N weight-shape coverage.
  • Validate Metal library existence, size, and SHA-256 binding.
  • Fail explicitly on missing or mismatched INT8 resources.
  • Keep the existing BF16 and other quantization paths unchanged.
  • Add focused tests for Qwen INT8 Metal binding validation.

Runtime Contract

The INT8 model must provide the matching apple_silicon_metal/ package and precompiled Metal libraries. The runtime must not silently fall back to BF16 when the INT8 Metal dependency is invalid.

Follow-up

After upstream merge, the runtime should be packaged and published as an official release artifact for the macOS App to consume.

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