Models trained with PARQ: https://arxiv.org/abs/2503.15748 using TorchAO
pytorch
non-profit
AI & ML interests
An open source machine learning framework that accelerates the path from research prototyping to production deployment.
Recent Activity
View all activity
-
pytorch/gemma-3-12b-it-FP8
Image-Text-to-Text • Updated • 1.62k • 1 -
pytorch/gemma-3-12b-it-INT4
Image-Text-to-Text • 3B • Updated • 1.02k -
pytorch/gemma-3-12b-it-AWQ-INT4
Image-Text-to-Text • Updated • 1.33k • 1 -
pytorch/gemma-3-27b-it-FP8
Image-Text-to-Text • Updated • 1.37k • 3
TorchAO quantized Phi-4-mini-instruct models from PyTorch team, runnable in A100, H100 through vLLM and in mobile devices through ExecuTorch
-
pytorch/Phi-4-mini-instruct-INT4
Text Generation • Updated • 25 -
pytorch/Phi-4-mini-instruct-FP8
Text Generation • Updated • 379 • 1 -
pytorch/Phi-4-mini-instruct-INT8-INT4
Text Generation • Updated • 1.81k • 2 -
pytorch/Phi-4-mini-instruct-AWQ-INT4
Text Generation • Updated • 534 • 3
Models trained with PARQ: https://arxiv.org/abs/2503.15748 using TorchAO
models quantized through quantization aware training from torchao
-
pytorch/gemma-3-12b-it-FP8
Image-Text-to-Text • Updated • 1.62k • 1 -
pytorch/gemma-3-12b-it-INT4
Image-Text-to-Text • 3B • Updated • 1.02k -
pytorch/gemma-3-12b-it-AWQ-INT4
Image-Text-to-Text • Updated • 1.33k • 1 -
pytorch/gemma-3-27b-it-FP8
Image-Text-to-Text • Updated • 1.37k • 3
TorchAO quantized Qwen3 models from PyTorch team, runnable in A100, H100 through vLLM and in mobile devices through ExecuTorch
TorchAO quantized Phi-4-mini-instruct models from PyTorch team, runnable in A100, H100 through vLLM and in mobile devices through ExecuTorch
-
pytorch/Phi-4-mini-instruct-INT4
Text Generation • Updated • 25 -
pytorch/Phi-4-mini-instruct-FP8
Text Generation • Updated • 379 • 1 -
pytorch/Phi-4-mini-instruct-INT8-INT4
Text Generation • Updated • 1.81k • 2 -
pytorch/Phi-4-mini-instruct-AWQ-INT4
Text Generation • Updated • 534 • 3