Download Latest Version Version 2.11.0 source code.tar.gz (8.0 MB)
Email in envelope

Get an email when there's a new version of AIMET

Home / 2.8.0
Name Modified Size InfoDownloads / Week
Parent folder
aimet_onnx-2.8.0+cpu-cp310-cp310-manylinux_2_34_x86_64.whl 2025-06-18 998.4 kB
0
aimet_onnx-2.8.0+cu121-cp310-cp310-manylinux_2_34_x86_64.whl 2025-06-18 33.4 MB
0
aimet_tf-2.8.0+cpu-cp310-cp310-manylinux_2_34_x86_64.whl 2025-06-18 5.5 MB
0
aimet_tf-2.8.0+cu118-cp310-cp310-manylinux_2_34_x86_64.whl 2025-06-18 16.2 MB
0
aimet_torch-2.8.0+cpu-py38-none-any.whl 2025-06-18 1.4 MB
0
aimet_torch-2.8.0+cu121-py38-none-any.whl 2025-06-18 9.4 MB
0
README.md 2025-06-18 1.3 kB
0
Version 2.8.0 source code.tar.gz 2025-06-18 7.4 MB
0
Version 2.8.0 source code.zip 2025-06-18 8.8 MB
0
Totals: 9 Items   83.1 MB 0

What's Changed

New Features

  • ONNX
    • Update aimet_onnx QuantizationSimModel.__init__ function signature (cbe67ae)
    • Defined new AdaRound API aimet_onnx.apply_adaround (84edcf5)
    • Defined new sequential MSE API aimet_onnx.apply_seq_mse (836ab1e)
    • Defined new per-layer sensitivity analysis API aimet_onnx.analyze_per_layer_sensitivity (dc34fa4)
    • Allowed onnx QuantizationSimModel.compute_encodings to take iterables (2c8ae88)
  • PyTorch
    • Added native support for huggingface Phi-3 (80cd141)

Bug Fixes and Improvements * ONNX * Made dynamic weights of Conv, ConvTranspose, Gemm, and MatMul follow the symmetry of static weights (ce68e75) * aimet-onnx on PyPI is now compatible with onnxruntime-gpu (6d3aa97) * Unpinned onnx version (abe8782) * Changed default execution provider to CPUExecutionProvider (e7d10c7) * Made QcQuantizeOp's data_type attribute always consistent without additional reconfiguration (8009871) * Made delta/offset and min/max always consistent (88706ef) * PyTorch * Made input quantizers always get enabled whenever the input wasn't already quantized (a2adae2) * Deprecated saving PyTorch model object during QuantizationsimModel.export (b5521f3)

Source: README.md, updated 2025-06-18