Skip to content

[Draft] Winml serve - #314

Closed
Qiong Wu (qiowu) (DingmaomaoBJTU) wants to merge 3 commits into
mainfrom
qiowu/wmk_serve
Closed

Qiong Wu (qiowu) (DingmaomaoBJTU) wants to merge 3 commits into
mainfrom
qiowu/wmk_serve

Conversation

@DingmaomaoBJTU

Copy link
Copy Markdown
Collaborator

No description provided.

@DingmaomaoBJTU Qiong Wu (qiowu) (DingmaomaoBJTU) changed the title Qiowu/wmk serve [Draft] Winml serve Apr 12, 2026
click.echo(f"Error during inference: {exc}", err=True)
sys.exit(4)

_print_result(result.model_dump(), output_format=output_format, output_path=output)
except ImportError as e:
click.echo(f"Error: Failed to load serving module: {e}", err=True)
sys.exit(1)
_banner0(host=host, port=port)
click.echo(f"Error: Failed to load serving module: {e}", err=True)
sys.exit(1)
_banner0(host=host, port=port)
uvicorn.run(app, host=host, port=port, reload=reload, log_level="warning")
click.echo(f"Error: Failed to load serving module: {e}", err=True)
sys.exit(1)
_banner0(host=host, port=port)
uvicorn.run(app, host=host, port=port, reload=reload, log_level="warning")
sys.exit(1)

mode = "multi" if multi else "single"
inference_app = create_app(
description=description,
)
self._slots[model_id] = slot
logger.info("ModelSlotManager: loaded %s (alias=%s)", model_id, alias)
logger.info("ModelSlotManager: idle timeout — evicting %s", model_id)
slot.engine.unload()
del self._slots[model_id]
except asyncio.CancelledError:
self._idle_timeout,
)
self._engine.unload()
except asyncio.CancelledError:

async def list_models(self) -> list[dict]:
"""Return metadata for all currently registered/loaded models."""
...
Comment thread src/winml/modelkit/serve/manager.py Dismissed
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.

3 participants