Search results

Module-wise type hinting and docstring coverage audit for PT2E

Status: Open.

#4541 In pytorch/ao;

· GiGiKoneti opened on Jun 28, 2026

Add scale and zero-point dimension validation to low-level quantization primitives

Status: Open.

#4538 In pytorch/ao;

· GiGiKoneti opened on Jun 27, 2026

Question: Lowering a custom TorchAO QAT model to XNNPACK backend

Status: Open.

#4529 In pytorch/ao;

· Lorenzo-Mazza opened on Jun 23, 2026

Linear+BatchNorm1d not folded during QAT, unlike Conv+BN

enhancementNew feature or requestNew feature or request

Status: Open.

#4463 In pytorch/ao;

· kreefduik opened on Jun 4, 2026

convert_pt2e crashes in DuplicateDQPass — GraphModule rebuild drops submodules only referenced by other submodules' internal graphs

bugSomething isn't workingSomething isn't working

Status: Open.

#4455 In pytorch/ao;

· ThreeD-H opened on May 29, 2026

Float8DynamicActivationFloat8WeightConfig + KernelPreference.MSLK + torch.compile fails on torchao 0.17.0 (missing import mslk.gemm in float8_tensor.py)

Status: Open.

#4438 In pytorch/ao;

· WaterKnight1998 opened on May 25, 2026

constant_fold deletes shared get_attr target while another get_attr for the same target is still live

bugSomething isn't workingSomething isn't working

module: pt2e_quantpt2 export quantization (prepare_pt2e, convert_pt2e, quantizer)pt2 export quantization (prepare_pt2e, convert_pt2e, quantizer)

Status: Open.

#4420 In pytorch/ao;

· AdrianLundell opened on May 18, 2026

[RFC] MoE Quantization-Aware Training (QAT) in TorchAO

Status: Open.

#4408 In pytorch/ao;

· CaoSY opened on May 15, 2026

Lightning Supply Chain Attack -- Potential risk via torch.ao?

Status: Open.

#4385 In pytorch/ao;

· codingbutstillalive opened on May 12, 2026

Float8DynamicActivationFloat8WeightConfig is inference-only — backward fails on torch._scaled_mm; document or add training-aware variant

Status: Open.

#4376 In pytorch/ao;

· genno-whittlery opened on May 11, 2026

Int8DynamicActivationInt8WeightConfig noticeably slower than torch.ao's quantize_dynamic()

cpu

performance

Status: Open.

#4365 In pytorch/ao;

· mscdex opened on May 5, 2026

[RFC] Enabling MoE Inference Quantization in TorchAO

tracker

Status: Open.

#4355 In pytorch/ao;

· xiaowangintel opened on Apr 30, 2026