pycolmap/_core.cpython-310-x86_64-linux-gnu.so error.How to solve this problem

Basic package list:

uv pip list
Package Version Editable project location
------------------------- --------------- ----------------------------------------------------------
addict 2.4.0
annotated-doc 0.0.4
annotated-types 0.7.0
antlr4-python3-runtime 4.9.3
anyio 4.12.1
argcomplete 3.6.3
attrs 25.4.0
blinker 1.9.0
certifi 2026.1.4
cfgv 3.5.0
charset-normalizer 3.4.4
click 8.3.1
colorama 0.4.6
configargparse 1.7.1
contourpy 1.3.2
cuda-bindings 12.9.4
cuda-pathfinder 1.3.3
cycler 0.12.1
dash 3.4.0
decorator 4.4.2
depth-anything-3 0.0.0 /workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main
distlib 0.4.0
e3nn 0.5.9
einops 0.8.2
evo 1.34.2
exceptiongroup 1.3.1
fastapi 0.128.0
fastjsonschema 2.21.2
filelock 3.20.3
flask 3.1.2
fonttools 4.61.1
fsspec 2026.1.0
gsplat 1.5.3+pt24cu124
h11 0.16.0
hf-xet 1.2.0
httpcore 1.0.9
httpx 0.28.1
huggingface-hub 1.3.5
identify 2.6.16
idna 3.11
imageio 2.37.2
imageio-ffmpeg 0.6.0
importlib-metadata 8.7.1
itsdangerous 2.2.0
jaxtyping 0.3.6
jinja2 3.1.6
joblib 1.5.3
jsonschema 4.26.0
jsonschema-specifications 2025.9.1
jupyter-core 5.9.1
kiwisolver 1.4.9
lz4 4.4.5
markdown-it-py 4.0.0
markupsafe 3.0.3
matplotlib 3.10.8
mdurl 0.1.2
moviepy 1.0.3
mpmath 1.3.0
narwhals 2.15.0
natsort 8.4.0
nbformat 5.10.4
nest-asyncio 1.6.0
networkx 3.4.2
ninja 1.13.0
nodeenv 1.10.0
numexpr 2.14.1
numpy 1.26.4
nvidia-cublas-cu12 12.8.4.1
nvidia-cuda-cupti-cu12 12.8.90
nvidia-cuda-nvrtc-cu12 12.8.93
nvidia-cuda-runtime-cu12 12.8.90
nvidia-cudnn-cu12 9.10.2.21
nvidia-cufft-cu12 11.3.3.83
nvidia-cufile-cu12 1.13.1.3
nvidia-curand-cu12 10.3.9.90
nvidia-cusolver-cu12 11.7.3.90
nvidia-cusparse-cu12 12.5.8.93
nvidia-cusparselt-cu12 0.7.1
nvidia-nccl-cu12 2.27.5
nvidia-nvjitlink-cu12 12.8.93
nvidia-nvshmem-cu12 3.4.5
nvidia-nvtx-cu12 12.8.90
omegaconf 2.3.0
open3d 0.19.0
opencv-python 4.11.0.86
opt-einsum 3.4.0
opt-einsum-fx 0.1.4
packaging 26.0
pandas 2.3.3
pathspec 1.0.4
pillow 12.1.0
pillow-heif 1.2.0
platformdirs 4.5.1
plotly 6.5.2
plyfile 1.1.3
pre-commit 4.5.1
proglog 0.1.12
pybind11 3.0.1
pycolmap 3.13.0
pydantic 2.12.5
pydantic-core 2.41.5
pygments 2.19.2
pyparsing 3.3.2
pyquaternion 0.9.9
python-dateutil 2.9.0.post0
pytz 2025.2
pyyaml 6.0.3
referencing 0.37.0
requests 2.32.5
retrying 1.4.2
rich 14.3.1
rosbags 0.11.0
rpds-py 0.30.0
ruamel-yaml 0.19.1
safetensors 0.7.0
scikit-build-core 0.11.6
scikit-learn 1.7.2
scipy 1.15.3
seaborn 0.13.2
setuptools 80.10.2
shellingham 1.5.4
six 1.17.0
starlette 0.50.0
sympy 1.14.0
threadpoolctl 3.6.0
tomli 2.4.0
torch 2.10.0
torchvision 0.25.0
tqdm 4.67.1
traitlets 5.14.3
trimesh 4.11.1
triton 3.6.0
typer 0.21.1
typer-slim 0.21.1
typing-extensions 4.15.0
typing-inspection 0.4.2
tzdata 2025.3
urllib3 2.6.3
uvicorn 0.40.0
virtualenv 20.36.1
wadler-lindig 0.1.7
werkzeug 3.1.5
xformers 0.0.34
zipp 3.23.0
zstandard 0.25.0
(base) root@devpod2:/workspace/nvidia-cosmos/Depth-Anything-3-main#

uv run --python python3.10 python -c "import torch; print(torch.version.cuda); print(torch.backends.cudnn.version()); print(torch.__version__)"
12.8
91002
2.10.0+cu128

2, The error:
uv run --python python3.10 da3 auto assets/examples/SOH \
--export-format glb \
--export-dir ./out \
--model-dir /workspace/huggingface/hub/depth-anything/DA3NESTED-GIANT-LARGE-1.1
🔍 Detected input type: IMAGES
📁 Input path: assets/examples/SOH

Processing directory of images...
Found 2 images to process
Export directory './out' already exists.
Do you want to clean it and continue? [y/N]: y
Cleaned export directory: ./out
Loading model from /workspace/huggingface/hub/depth-anything/DA3NESTED-GIANT-LARGE-1.1...
[INFO ] using SwiGLU layer as FFN
[INFO ] using MLP layer as FFN
Loading weights from local directory
Running inference on 2 images...
[INFO ] Processed Images Done taking 0.05663895606994629 seconds. Shape: torch.Size([2, 3, 280, 504])
*** Aborted at 1769772423 (unix time) try "date -d @1769772423" if you are using GNU date ***
PC: @ 0x7fe48836f137 (/workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main/.venv/lib/python3.10/site-packages/nvidia/cublas/lib/libcublas.so.12+0x16f136)
*** SIGFPE (@0x7fe48836f137) received by PID 1404504 (TID 0x7fe5321eb740) from PID 18446744071699886391; stack trace: ***
@ 0x7fe53228fed3 (/usr/lib/x86_64-linux-gnu/libc.so.6+0xa1ed2)
@ 0x7fe2cfdc6868 (/workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main/.venv/lib/python3.10/site-packages/pycolmap/_core.cpython-310-x86_64-linux-gnu.so+0xa26867)
@ 0x7fe532233330 (/usr/lib/x86_64-linux-gnu/libc.so.6+0x4532f)
@ 0x7fe48836f137 (/workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main/.venv/lib/python3.10/site-packages/nvidia/cublas/lib/libcublas.so.12+0x16f136)
@ 0x7fe48847e4b3 (/workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main/.venv/lib/python3.10/site-packages/nvidia/cublas/lib/libcublas.so.12+0x27e4b2)
@ 0x7fe48847efec (/workspace/huggingface/nvidia-cosmos/Depth-Anything-3-main/.venv/lib/python3.10/site-packages/nvidia/cublas/lib/libcublas.so.12+0x27efeb)
@ 0x7fe4883293ae cublasSdot_v2
@ 0x7fe388de437f void at::cuda::blas::dot(cublasContext*, int, float const*, int, float const*, int, float*)
@ 0x7fe388e87f9a at::native::dot_cuda(at::Tensor const&, at::Tensor const&) [clone .localalias]
@ 0x7fe388add890 c10::impl::wrap_kernel_functor_unboxed_, at::Tensor, c10::g4
@ 0x7fe3c54ef481 at::_ops::dot::call(at::Tensor const&, at::Tensor const&)
@ 0x7fe3c4471d33 at::autocast::WrapFunction_<(at::autocast::CastPolicy)4, (c10::DeviceType)1, at::Tensor (at::Tensor const&, at::Tensor const&), &at::_ops::dot::call, at::Tensor, c10::guts::typelist::typelist >::call(at::Tensor const&,4
@ 0x7fe3c433e880 c10::impl::wrap_kernel_functor_unboxed_ >, at::Tensor (at::Tensor const4
@ 0x7fe3c54ef481 at::_ops::dot::call(at::Tensor const&, at::Tensor const&)
@ 0x7fe3d835e6d2 torch::autograd::THPVariable_dot(_object*, _object*, _object*)
@ 0x5612c5459162 cfunction_call
@ 0x5612c54a02fc _PyEval_EvalFrameDefault
@ 0x5612c5437bd2 _PyFunction_Vectorcall
@ 0x5612c54c4559 call_function
@ 0x5612c54915c1 _PyEval_EvalFrameDefault
@ 0x5612c5437bd2 _PyFunction_Vectorcall
@ 0x5612c5439bd0 method_vectorcall
@ 0x5612c54977c4 _PyEval_EvalFrameDefault
@ 0x5612c5437bd2 _PyFunction_Vectorcall
@ 0x5612c5439c6f method_vectorcall
@ 0x5612c54958f3 _PyEval_EvalFrameDefault
@ 0x5612c5437bd2 _PyFunction_Vectorcall
@ 0x5612c5439c6f method_vectorcall
@ 0x5612c54958f3 _PyEval_EvalFrameDefault
@ 0x5612c5437bd2 _PyFunction_Vectorcall
@ 0x5612c55364a6 _PyObject_Call_Prepend
@ 0x5612c5569fb6 slot_tp_call

3:,env:
(base) root@devpod2:/workspace/nvidia-cosmos/Depth-Anything-3-main# vidia-smi
Fri Jan 30 11:27:23 2026
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 550.144.03 Driver Version: 550.144.03 CUDA Version: 12.4 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA A100-PCIE-40GB Off | 00000000:31:00.0 Off | 0 |
| N/A 30C P0 36W / 250W | 1443MiB / 40960MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+------------------------+----------------------+
| 1 NVIDIA A100-PCIE-40GB Off | 00000000:4B:00.0 Off | 0 |
| N/A 29C P0 32W / 250W | 4MiB / 40960MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+------------------------+----------------------+
| 2 NVIDIA A100-PCIE-40GB Off | 00000000:B1:00.0 Off | 0 |
| N/A 29C P0 32W / 250W | 4MiB / 40960MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+------------------------+----------------------+
| 3 NVIDIA A100-PCIE-40GB Off | 00000000:CA:00.0 Off | 0 |
| N/A 30C P0 36W / 250W | 4MiB / 40960MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|

Start by reproducing the reported `uv run --python python3.10 da3 auto` command with the listed model and image directory. Trace the failure through `pycolmap/_core.cpython-310-x86_64-linux-gnu.so` and the PyTorch CUDA `dot` call, comparing the reported Python, PyTorch, CUDA, driver, and GPU versions. Done means the example command completes without the SIGFPE, with the compatible environment or required project change documented.

Written by the indexing model from the issue text.