Currently we only support a100 and t4 gpus. We may support other gpus in the future

[appleatiger]

When I enter "stochasticx stable-diffusion deploy --type aitemplate," the following message appears: "WARNING | stochasticx.client.stable_diffusion | Currently we only support a100 and t4 gpus. We may support other gpus in the future." My GPU is an RTX4090. Please add it to the list of supported devices.

[Toan-Do]

Hi @appleatiger, we have supported for more gpus like rtx 3080, 3090, 4080, 4090,... Please reinstall our latest stochasticx library , try again and let us know if any issue. Thank you.

[appleatiger]

"pip uninstall stochasticx" then "pip install stochasticx","stochasticx login","stochasticx stable-diffusion deploy --type aitemplate"
a few minute later ,it shows
"2023-01-05 09:42:02,949 | ERROR | stochasticx.utils.docker | 500 Server Error for http+docker://localnpipe/v1.41/containers/331de07878ab9cca10e9b8104bdd90021fd6804cf2bcf6ddafb5ba729c1b9bc5/start: Internal Server Error ("failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: mount error: file creation failed: /var/lib/docker/overlay2/90bf537a32793e05ef4f026edae56918ebf2b1052ee602c13bf8a99572adde6d/merged/usr/lib/x86_64-linux-gnu/libnvidia-ml.so.1: file exists: unknown")"

a docker image "public.ecr.aws/t8g5g2q5/stable-diffusion:aitemplate " and a container "stochasticx_stable_diffusion" is created。

[Toan-Do]

Hi @appleatiger, have you install nvidia-runtime in your machine and set the docker default runtime is nvidia-runtime ?