Skip to content

Conversation

chinazhangchao
Copy link
Contributor

No description provided.

@chinazhangchao chinazhangchao changed the title Chao/rtx Update winml version, add llm trt rtx recipes. Jul 17, 2025
"builder": { "type": "ModelBuilder", "precision": "fp16", "search": { "max_length": 2048, "min_length": 0 } },
"metadata": {
"type": "ModelBuilder",
"precision": "fp16",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

shall we do this? as amol said today, TRT should use QDQ

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

let's wait for nvidia

@@ -0,0 +1,42 @@
{
"name": "Convert to NVIDIA GPU (TRT RTX)",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Convert to NVIDIA TRT for RTX

" site_packages_path = Path(str(metadata.distribution('winrt-runtime').locate_file('')))\n",
" dll_path = site_packages_path / 'winrt' / 'msvcp140.dll'\n",
" if dll_path.exists():\n",
" dll_path.unlink()\n",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could we unload it once we get ep and before register instead of deletion?

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.

2 participants