search

stochasticai / x-stable-diffusion

Real-time inference for Stable Diffusion - 0.88s latency. Covers AITemplate, nvFuser, TensorRT, FlashAttention. Join our Discord communty: https://discord.com/invite/TgHXuSJEk6
https://stochastic.ai
Apache License 2.0
553 stars 35 forks source link

Currently we only support a100 and t4 gpus. We may support other gpus in the future #29

Open appleatiger opened 1 year ago

appleatiger commented 1 year ago

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 commented 1 year ago

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 commented 1 year ago

"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 commented 1 year ago

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