Hugging Face's logo
Ilyas Moutawwakil's picture
40 36 64

Ilyas Moutawwakil

IlyasMoutawwakil
m-ric's profile picture lysandre's profile picture reid765's profile picture
Ā·

AI & ML interests

Optimization, LLMs, Hardware, Backends, ..

Recent Activity

replied to their post about 6 hours ago
After 2 months of refinement, I'm happy to announce that a lot of Transformers' modeling code is now significantly more torch-compile & export-friendly šŸ”„ Why it had to be done šŸ‘‡ PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph ! Transformers models are now easier to: āš™ļø Compile end-to-end with torch.compile backends šŸ“¦ Export reliably via torch.export and torch.onnx.export šŸš€ Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes. This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators. We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere. There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs. PR in the comments ! More updates coming coming soon !
posted an update about 6 hours ago
After 2 months of refinement, I'm happy to announce that a lot of Transformers' modeling code is now significantly more torch-compile & export-friendly šŸ”„ Why it had to be done šŸ‘‡ PyTorch's Dynamo compiler is increasingly becoming the default interoperability layer for ML systems. Anything that relies on torch.export or torch.compile, from model optimization to cross-framework integrations, benefits directly when models can be captured as a single dynamo-traced graph ! Transformers models are now easier to: āš™ļø Compile end-to-end with torch.compile backends šŸ“¦ Export reliably via torch.export and torch.onnx.export šŸš€ Deploy to ONNX / ONNX Runtime, Intel Corporation's OpenVINO, NVIDIA AutoDeploy (TRT-LLM), AMD's Quark, Meta's Executorch and more hardware-specific runtimes. This work aims at unblocking entire TorchDynamo-based toolchains that rely on exporting Transformers across runtimes and accelerators. We are doubling down on Transformers commitment to be a first-class citizen of the PyTorch ecosystem, more exportable, more optimizable, and easier to deploy everywhere. There are definitely some edge-cases that we still haven't addressed so don't hesitate to try compiling / exporting your favorite transformers and to open issues / PRs. PR in the comments ! More updates coming coming soon !
liked a Space 8 days ago
nvidia/kvpress-leaderboard
View all activity

Organizations

Hugging Face's profile picture Training Transformers Together's profile picture OpenVINO Toolkit's profile picture Hugging Face Optimum's profile picture ONNXConfig for all's profile picture HugGAN Community's profile picture HuggingFaceM4's profile picture Hugging Face H4's profile picture Optimum AMD's profile picture That Time I got Reincarnated as a Hugging Face Organization's profile picture AI Energy Score's profile picture Optimum-Benchmark's profile picture Social Post Explorers's profile picture Dev Mode Explorers's profile picture Optimum-Intel's profile picture ONNX Community's profile picture Hugging Face Machine Learning Optimization's profile picture Optimum Internal Testing's profile picture CI org's profile picture Optimum Intel Internal Testing's profile picture