CUDA on GTX Titan Black

[jemand.exempels]

Hi,

I recently got a hold of 2 GTX Titan Black’s and I was trying to get CUDA running on them.
I’m trying to use them for playing around with ai/ml in Python.
But as of now I couldn’t get CUDA to work on this PC.
I’m using Driver version 470.223.02 with CUDA version 11.4 and Ubuntu 22.04.
When trying to download the CUDA toolkit for that version the latest Ubuntu version supported is 20.04.
Is there any way to get CUDA running on this system without reinstalling Ubuntu (downgrading to 20.04)?

Thanks in advance.

[c_cool_la_vie]

Dear friend,

I just had a long issue, solved with the fking master Claude Opus on Vscode, and published this post on the topic + link to recompiled pytorch, gcc8, with python 3.10.

• **NVIDIA Driver : ** R470 legacy, the last to still support Kepler architecture) — to download from NVIDIA.
• **The CUDA 10.2 toolkit ** (~2,5 Go, with extras/CUPTI) — to download from official NVIDIA Archive (developer.nvidia.com/cuda-10.2-download-archive).

Check this complete post :

Installing Pytorch with Python 3.10 for CUDA on NVIDIA TITAN BLACK

Hello all. I have spent a little time, helped with Claude in VS Code, connected through a WSL Ubuntu on my Windows machine, to have my old NVIDIA TITAN BLACK work well. Python 3 didn’t work well, gcc compiler version was too recent, lots of problems that Claude mastered through 4-5 days of debugging, one command at a time after my current work. If useful for anyone, here is the result successful config. By the way, a TAR upload for the project including compiled whl is available here : mega.…

Working version matrix

Component	Version	Notes
GPU	GTX Titan Black	Kepler, compute capability 3.5
Driver	470.256.02	R470 legacy branch — last to support Kepler. Do not upgrade.
OS	Ubuntu 24.04 (noble), WSL2
CUDA toolkit	10.2	Installed via runfile, not apt
Host compiler	GCC 8.5.0	Via conda-forge (miniforge), no root needed
Linker	System ld 2.42	GCC8-conda’s own ld (2.36.1) too old for RELR relocations in glibc 2.39
Python	3.10.20	venv, PyTorch 1.12.1 doesn’t support 3.12 (Ubuntu 24.04 default)
cmake	3.22.6	Newer cmake 4.x breaks on protobuf’s old cmake_minimum_required
ninja	1.13.0
numpy (build+runtime)	2.2.6	Needed 1-line source patch, see below
PyTorch source	tag v1.12.1, commit 664058f	USE_NCCL=0, USE_DISTRIBUTED=0 (single GPU, NCCL build system ignores host compiler override)

Output: torch-1.12.0a0+git664058f-cp310-cp310-linux_x86_64.whl, ~159MB, 9925 files.

If it helps, get it directly here (bonus : it’s on ubuntu 24.04 ! )
mega.nz/file/BcsHEKab
With key = 6_9bGyOYvDOEoxmV9QXNwZvw0lJUTkOUMG2HppvWSYo

Checksum :

sha256sum torch-1.12.0a0+git664058f-cp310-cp310-linux_x86_64.whl
# has to result to : 1e8b423910dcf33b9286f27180066aeda0d1dda1ae2a0e0ebb273c1ce4b1c931