Issues

• aten::searchsorted lowering drops every keyword argument (right, side, out_int32, sorter)

  google/torchax#106 · Filter by author jt226ub opened on Sep 8, 2026

• avg_pool on integer inputs lowers through float32

  google/torchax#105 · Filter by author benSepanski opened on Sep 3, 2026
  1

• torch named tensors API (Dimname etc.) incompatible with torch >= 2.13

  google/torchax#101 · Filter by author ocramz opened on Jul 9, 2026
  1

  

• How to integrate Torch model with Flax nnx?

  google/torchax#94 · Filter by author jorgenin opened on May 11, 2026
  1

• How torchax handle tied_embedding?

  google/torchax#68 · Filter by author carlesoctav opened on Jan 22, 2026
  1

• train_llama_torchtitan example: Multiple issues running on TPU v4-32

  google/torchax#67 · Filter by author KarthikRevanuru opened on Jan 21, 2026
  2

• Convolution backward op missing

  google/torchax#65 · Filter by author samos123 opened on Jan 14, 2026
  1

• Unable to export quantized models to stablehlo

  google/torchax#62 · Filter by author Ha-Eyz opened on Dec 18, 2025
  2

• Torchax vs Torch2Jax ?

  google/torchax#55 · Filter by author adam-hartshorne opened on Nov 25, 2025
  3

• Dependency issue of 0.0.9

  google/torchax#30 · Filter by author wdhongtw opened on Nov 7, 2025
  1

• Buffers not moving to jax device

  google/torchax#23 · Filter by author mitkotak opened on Oct 29, 2025
  1

• Would nn.BatchNorm2d work as expected?

  google/torchax#22 · Filter by author lime-j opened on Oct 26, 2025
  13

• How do RNGs work in torchax

  google/torchax#17 · Filter by author carlesoctav opened on Oct 24, 2025
  1

• GPU out of memory during large model transfer to Jax device

  google/torchax#16 · Filter by author mozuysal-hubx opened on Oct 23, 2025
  4

• OperatorNotFound: Operator with name aten::_nested_tensor_from_mask_left_aligned has no lowering

  google/torchax#12 · Filter by author BioGeek opened on Oct 17, 2025
  3