Issue Station
Sign In
Tools hereToolbox

pytorch - 💡(How to fix) Fix [Bug] `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

torch.library.Library._register_fake calls torch._library.utils.get_source, which uses inspect.getframeinfo → findsource → getmodule purely to build a "<file>:<lineno>" diagnostic string. inspect.getmodule iterates sys.modules and does hasattr(module, "__file__") on every entry. For modules registered via importlib.util.LazyLoader this triggers LazyLoader.__getattribute__, which fully executes the lazy module body. If that body raises (e.g. ImportError for a missing optional dependency), the exception propagates out of an otherwise unrelated _register_fake call.

The op being registered has nothing to do with the lazy module. The lazy module is not on the call stack. Yet registration fails.

This is observable in the wild: it crashes vLLM at startup when cupy is installed but pytest is not (see vllm-project/vllm#43480). cupy registers cupy.testing as a LazyLoader placeholder; cupy/testing/_random.py does import pytest; any first-time _register_fake after cupy is imported then fails with ModuleNotFoundError: No module named 'pytest' from inside inspect.getmodule.

Error Message

File ".../torch/library.py", line 182, in _register_fake source = torch._library.utils.get_source(_stacklevel + 1) File ".../torch/_library/utils.py", line 45, in get_source frame = inspect.getframeinfo(sys._getframe(stacklevel)) File ".../python3.12/inspect.py", line 1844, in getframeinfo lines, lnum = findsource(frame) File ".../python3.12/inspect.py", line 1139, in findsource module = getmodule(object, file) File ".../python3.12/inspect.py", line 1058, in getmodule if ismodule(module) and hasattr(module, "file"): File "<frozen importlib.util>", line 207, in getattribute File ".../boom.py", line 1, in <module> assert False, "boom: lazy module was forced to load by inspect.getmodule" AssertionError: boom: lazy module was forced to load by inspect.getmodule

Root Cause

torch.library.Library._register_fake calls torch._library.utils.get_source, which uses inspect.getframeinfo → findsource → getmodule purely to build a "<file>:<lineno>" diagnostic string. inspect.getmodule iterates sys.modules and does hasattr(module, "__file__") on every entry. For modules registered via importlib.util.LazyLoader this triggers LazyLoader.__getattribute__, which fully executes the lazy module body. If that body raises (e.g. ImportError for a missing optional dependency), the exception propagates out of an otherwise unrelated _register_fake call.

The op being registered has nothing to do with the lazy module. The lazy module is not on the call stack. Yet registration fails.

This is observable in the wild: it crashes vLLM at startup when cupy is installed but pytest is not (see vllm-project/vllm#43480). cupy registers cupy.testing as a LazyLoader placeholder; cupy/testing/_random.py does import pytest; any first-time _register_fake after cupy is imported then fails with ModuleNotFoundError: No module named 'pytest' from inside inspect.getmodule.

Fix Action

Fix / Workaround

Step 2: register a custom op with a fake kernel. The op has nothing

to do with boom; boom is not on the call stack.

lib = torch.library.Library("repro", "FRAGMENT") lib.define("noop(Tensor x) -> Tensor") lib.impl("noop", lambda x: x.clone(), dispatch_key="CPU") lib._register_fake("noop", lambda x: torch.empty_like(x))


CPU:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   46 bits physical, 48 bits virtual
CPU(s):                          28
On-line CPU(s) list:             0-27
Thread(s) per core:              2
Core(s) per socket:              14
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           85
Model name:                      Intel(R) Core(TM) i9-10940X CPU @ 3.30GHz
Stepping:                        7
CPU MHz:                         4206.468
CPU max MHz:                     4800.0000
CPU min MHz:                     1200.0000
BogoMIPS:                        6599.98
Virtualization:                  VT-x
L1d cache:                       448 KiB
L1i cache:                       448 KiB
L2 cache:                        14 MiB
L3 cache:                        19.3 MiB
NUMA node0 CPU(s):               0-27
Vulnerability Itlb multihit:     KVM: Mitigation: VMX disabled
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Retbleed:          Vulnerable
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Vulnerable, IBPB: disabled, STIBP: disabled, PBRSB-eIBRS: Vulnerable
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Mitigation; TSX disabled
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512_vnni md_clear flush_l1d arch_capabilities

Code Example

assert False, "boom: lazy module was forced to load by inspect.getmodule"

---

import sys
import importlib.util
import torch


def lazy_import(name):
    spec = importlib.util.find_spec(name)
    spec.loader = importlib.util.LazyLoader(spec.loader)
    mod = importlib.util.module_from_spec(spec)
    sys.modules[name] = mod
    spec.loader.exec_module(mod)
    return mod


# Step 1: register an unrelated module as a LazyLoader placeholder
# (this is exactly what cupy does with cupy.testing).
lazy_import("boom")

# Step 2: register a custom op with a fake kernel. The op has nothing
# to do with `boom`; `boom` is not on the call stack.
lib = torch.library.Library("repro", "FRAGMENT")
lib.define("noop(Tensor x) -> Tensor")
lib.impl("noop", lambda x: x.clone(), dispatch_key="CPU")
lib._register_fake("noop", lambda x: torch.empty_like(x))

---

$ python repro.py

---

File ".../torch/library.py", line 182, in _register_fake
    source = torch._library.utils.get_source(_stacklevel + 1)
File ".../torch/_library/utils.py", line 45, in get_source
    frame = inspect.getframeinfo(sys._getframe(stacklevel))
File ".../python3.12/inspect.py", line 1844, in getframeinfo
    lines, lnum = findsource(frame)
File ".../python3.12/inspect.py", line 1139, in findsource
    module = getmodule(object, file)
File ".../python3.12/inspect.py", line 1058, in getmodule
    if ismodule(module) and hasattr(module, "__file__"):
File "<frozen importlib.util>", line 207, in __getattribute__
File ".../boom.py", line 1, in <module>
    assert False, "boom: lazy module was forced to load by inspect.getmodule"
AssertionError: boom: lazy module was forced to load by inspect.getmodule

---

Collecting environment information...
PyTorch version: 2.10.0+cu128
Is debug build: False
CUDA used to build PyTorch: 12.8
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 11.5.0-10ubuntu1~20~ppa3) 11.5.0
Clang version: Could not collect
CMake version: version 4.3.1
Libc version: glibc-2.31

Python version: 3.12.11 | packaged by Anaconda, Inc. | (main, Jun  5 2025, 13:09:17) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-76-generic-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to:
GPU models and configuration:
GPU 0: NVIDIA GeForce RTX 4090
GPU 1: NVIDIA GeForce RTX 4090

Nvidia driver version: 570.133.20
cuDNN version: Could not collect
Is XPU available: False
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
Caching allocator config: N/A

CPU:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   46 bits physical, 48 bits virtual
CPU(s):                          28
On-line CPU(s) list:             0-27
Thread(s) per core:              2
Core(s) per socket:              14
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           85
Model name:                      Intel(R) Core(TM) i9-10940X CPU @ 3.30GHz
Stepping:                        7
CPU MHz:                         4206.468
CPU max MHz:                     4800.0000
CPU min MHz:                     1200.0000
BogoMIPS:                        6599.98
Virtualization:                  VT-x
L1d cache:                       448 KiB
L1i cache:                       448 KiB
L2 cache:                        14 MiB
L3 cache:                        19.3 MiB
NUMA node0 CPU(s):               0-27
Vulnerability Itlb multihit:     KVM: Mitigation: VMX disabled
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Retbleed:          Vulnerable
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Vulnerable, IBPB: disabled, STIBP: disabled, PBRSB-eIBRS: Vulnerable
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Mitigation; TSX disabled
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512_vnni md_clear flush_l1d arch_capabilities

Versions of relevant libraries:
[pip3] fused_moe_cuda2==0.0.1+torch2.8cu129
[pip3] mypy==1.19.1
[pip3] mypy_extensions==1.1.0
[pip3] numpy==2.1.0
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime==13.2.51
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.13.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] torch==2.10.0+cu128
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0+cu128
[pip3] triton==3.6.0
[pip3] triton_kernels==1.0.0
[conda] fused-moe-cuda2           0.0.1+torch2.8cu129          pypi_0    pypi
[conda] numpy                     2.1.0                    pypi_0    pypi
[conda] nvidia-cublas-cu12        12.8.4.1                 pypi_0    pypi
[conda] nvidia-cuda-cupti-cu12    12.8.90                  pypi_0    pypi
[conda] nvidia-cuda-nvrtc-cu12    12.8.93                  pypi_0    pypi
[conda] nvidia-cuda-runtime       13.2.51                  pypi_0    pypi
[conda] nvidia-cuda-runtime-cu12  12.8.90                  pypi_0    pypi
[conda] nvidia-cudnn-cu12         9.10.2.21                pypi_0    pypi
[conda] nvidia-cudnn-frontend     1.13.0                   pypi_0    pypi
[conda] nvidia-cufft-cu12         11.3.3.83                pypi_0    pypi
[conda] nvidia-curand-cu12        10.3.9.90                pypi_0    pypi
[conda] nvidia-cusolver-cu12      11.7.3.90                pypi_0    pypi
[conda] nvidia-cusparse-cu12      12.5.8.93                pypi_0    pypi
[conda] nvidia-cusparselt-cu12    0.7.1                    pypi_0    pypi
[conda] nvidia-nccl-cu12          2.27.5                   pypi_0    pypi
[conda] nvidia-nvjitlink-cu12     12.8.93                  pypi_0    pypi
[conda] nvidia-nvtx-cu12          12.8.90                  pypi_0    pypi
[conda] torch                     2.10.0+cu128             pypi_0    pypi
[conda] torch-c-dlpack-ext        0.1.5                    pypi_0    pypi
[conda] torchaudio                2.10.0                   pypi_0    pypi
[conda] torchvision               0.25.0+cu128             pypi_0    pypi
[conda] triton                    3.6.0                    pypi_0    pypi
[conda] triton-kernels            1.0.0                    pypi_0    pypi
RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

Summary

torch.library.Library._register_fake calls torch._library.utils.get_source, which uses inspect.getframeinfo → findsource → getmodule purely to build a "<file>:<lineno>" diagnostic string. inspect.getmodule iterates sys.modules and does hasattr(module, "__file__") on every entry. For modules registered via importlib.util.LazyLoader this triggers LazyLoader.__getattribute__, which fully executes the lazy module body. If that body raises (e.g. ImportError for a missing optional dependency), the exception propagates out of an otherwise unrelated _register_fake call.

The op being registered has nothing to do with the lazy module. The lazy module is not on the call stack. Yet registration fails.

This is observable in the wild: it crashes vLLM at startup when cupy is installed but pytest is not (see vllm-project/vllm#43480). cupy registers cupy.testing as a LazyLoader placeholder; cupy/testing/_random.py does import pytest; any first-time _register_fake after cupy is imported then fails with ModuleNotFoundError: No module named 'pytest' from inside inspect.getmodule.

Minimal reproduction

boom.py:

assert False, "boom: lazy module was forced to load by inspect.getmodule"

repro.py:

import sys
import importlib.util
import torch


def lazy_import(name):
    spec = importlib.util.find_spec(name)
    spec.loader = importlib.util.LazyLoader(spec.loader)
    mod = importlib.util.module_from_spec(spec)
    sys.modules[name] = mod
    spec.loader.exec_module(mod)
    return mod


# Step 1: register an unrelated module as a LazyLoader placeholder
# (this is exactly what cupy does with cupy.testing).
lazy_import("boom")

# Step 2: register a custom op with a fake kernel. The op has nothing
# to do with `boom`; `boom` is not on the call stack.
lib = torch.library.Library("repro", "FRAGMENT")
lib.define("noop(Tensor x) -> Tensor")
lib.impl("noop", lambda x: x.clone(), dispatch_key="CPU")
lib._register_fake("noop", lambda x: torch.empty_like(x))

Run from the directory containing both:

$ python repro.py

Actual output (relevant tail):

File ".../torch/library.py", line 182, in _register_fake
    source = torch._library.utils.get_source(_stacklevel + 1)
File ".../torch/_library/utils.py", line 45, in get_source
    frame = inspect.getframeinfo(sys._getframe(stacklevel))
File ".../python3.12/inspect.py", line 1844, in getframeinfo
    lines, lnum = findsource(frame)
File ".../python3.12/inspect.py", line 1139, in findsource
    module = getmodule(object, file)
File ".../python3.12/inspect.py", line 1058, in getmodule
    if ismodule(module) and hasattr(module, "__file__"):
File "<frozen importlib.util>", line 207, in __getattribute__
File ".../boom.py", line 1, in <module>
    assert False, "boom: lazy module was forced to load by inspect.getmodule"
AssertionError: boom: lazy module was forced to load by inspect.getmodule

Expected: _register_fake succeeds. boom is unrelated to the op and should not be executed.

Versions

Collecting environment information...
PyTorch version: 2.10.0+cu128
Is debug build: False
CUDA used to build PyTorch: 12.8
ROCM used to build PyTorch: N/A

OS: Ubuntu 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 11.5.0-10ubuntu1~20~ppa3) 11.5.0
Clang version: Could not collect
CMake version: version 4.3.1
Libc version: glibc-2.31

Python version: 3.12.11 | packaged by Anaconda, Inc. | (main, Jun  5 2025, 13:09:17) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-76-generic-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to:
GPU models and configuration:
GPU 0: NVIDIA GeForce RTX 4090
GPU 1: NVIDIA GeForce RTX 4090

Nvidia driver version: 570.133.20
cuDNN version: Could not collect
Is XPU available: False
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True
Caching allocator config: N/A

CPU:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   46 bits physical, 48 bits virtual
CPU(s):                          28
On-line CPU(s) list:             0-27
Thread(s) per core:              2
Core(s) per socket:              14
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           85
Model name:                      Intel(R) Core(TM) i9-10940X CPU @ 3.30GHz
Stepping:                        7
CPU MHz:                         4206.468
CPU max MHz:                     4800.0000
CPU min MHz:                     1200.0000
BogoMIPS:                        6599.98
Virtualization:                  VT-x
L1d cache:                       448 KiB
L1i cache:                       448 KiB
L2 cache:                        14 MiB
L3 cache:                        19.3 MiB
NUMA node0 CPU(s):               0-27
Vulnerability Itlb multihit:     KVM: Mitigation: VMX disabled
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Retbleed:          Vulnerable
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Vulnerable, IBPB: disabled, STIBP: disabled, PBRSB-eIBRS: Vulnerable
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Mitigation; TSX disabled
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512_vnni md_clear flush_l1d arch_capabilities

Versions of relevant libraries:
[pip3] fused_moe_cuda2==0.0.1+torch2.8cu129
[pip3] mypy==1.19.1
[pip3] mypy_extensions==1.1.0
[pip3] numpy==2.1.0
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime==13.2.51
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cudnn-frontend==1.13.0
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-nccl-cu12==2.27.5
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] torch==2.10.0+cu128
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.10.0
[pip3] torchvision==0.25.0+cu128
[pip3] triton==3.6.0
[pip3] triton_kernels==1.0.0
[conda] fused-moe-cuda2           0.0.1+torch2.8cu129          pypi_0    pypi
[conda] numpy                     2.1.0                    pypi_0    pypi
[conda] nvidia-cublas-cu12        12.8.4.1                 pypi_0    pypi
[conda] nvidia-cuda-cupti-cu12    12.8.90                  pypi_0    pypi
[conda] nvidia-cuda-nvrtc-cu12    12.8.93                  pypi_0    pypi
[conda] nvidia-cuda-runtime       13.2.51                  pypi_0    pypi
[conda] nvidia-cuda-runtime-cu12  12.8.90                  pypi_0    pypi
[conda] nvidia-cudnn-cu12         9.10.2.21                pypi_0    pypi
[conda] nvidia-cudnn-frontend     1.13.0                   pypi_0    pypi
[conda] nvidia-cufft-cu12         11.3.3.83                pypi_0    pypi
[conda] nvidia-curand-cu12        10.3.9.90                pypi_0    pypi
[conda] nvidia-cusolver-cu12      11.7.3.90                pypi_0    pypi
[conda] nvidia-cusparse-cu12      12.5.8.93                pypi_0    pypi
[conda] nvidia-cusparselt-cu12    0.7.1                    pypi_0    pypi
[conda] nvidia-nccl-cu12          2.27.5                   pypi_0    pypi
[conda] nvidia-nvjitlink-cu12     12.8.93                  pypi_0    pypi
[conda] nvidia-nvtx-cu12          12.8.90                  pypi_0    pypi
[conda] torch                     2.10.0+cu128             pypi_0    pypi
[conda] torch-c-dlpack-ext        0.1.5                    pypi_0    pypi
[conda] torchaudio                2.10.0                   pypi_0    pypi
[conda] torchvision               0.25.0+cu128             pypi_0    pypi
[conda] triton                    3.6.0                    pypi_0    pypi
[conda] triton-kernels            1.0.0                    pypi_0    pypi

cc @anjali411 @chauhang @penguinwu @bdhirsh @bobrenjc93 @aorenste

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

pytorch - 💡(How to fix) Fix [Bug] `torch.library._register_fake` crashes when any unrelated `LazyLoader` module in `sys.modules` fails to import