Search results

Training hangs after 1 epoch when using QAT

Status: Open.

#2423 In pytorch/ao;

· kiyoonyoo opened on Jun 22, 2025

Benefits of Using QAT Before GGUF Quantization?

Status: Open.

#2419 In pytorch/ao;

· kiyoonyoo opened on Jun 21, 2025

TorchAO Paper

Status: Open.

#2412 In pytorch/ao;

· FabianSchuetze opened on Jun 19, 2025

TP + FSDP + MXFP8 fails during compile

Status: Open.

#2393 In pytorch/ao;

· mreso opened on Jun 17, 2025

Implement an AWQ algorithm with dynamic activation quantization for ExecuTorch

Status: Open.

#2388 In pytorch/ao;

· metascroy opened on Jun 16, 2025

Support cpu wheel with cpp kernels

Status: Open.

#2378 In pytorch/ao;

· Valentine233 opened on Jun 16, 2025

bug: int8 w8a8 doesn't work on 5090

Status: Open.

Bug

#2376 In pytorch/ao;

· gau-nernst opened on Jun 14, 2025

Remove old subclass APIs are no longer need and are creating a maintenence burden

Status: Open.

#2368 In pytorch/ao;

· HDCharles opened on Jun 12, 2025

TorchAO ROCM tests are taking a long time

good first issueGood for newcomersGood for newcomers

Status: Open.

#2367 In pytorch/ao;

· HDCharles opened on Jun 12, 2025

Support <code>torch.int4</code> <code>target_dtype</code> for ops <code>choose_qparams_affine</code>, <code>quantize_affine</code>, <code>dequantize_affine</code>

Status: Open.

#2354 In pytorch/ao;

· u-simha opened on Jun 10, 2025

DISABLED test_int4_weight_only_quant_subclass_grouped_5_cuda (__main__.TestSubclass)

skipped

Status: Open.

#2352 In pytorch/ao;

· petrex opened on Jun 10, 2025

Add _apply_fn_to_data in AOBaseClass

good first issueGood for newcomersGood for newcomers

topic: new featureUse this tag if this PR adds a new featureUse this tag if this PR adds a new feature

Status: Open.

#2349 In pytorch/ao;

· drisspg opened on Jun 10, 2025