[GH-ISSUE #4] CUDA error: device-side assert triggered when trying to run the model #3

Closed
opened 2026-10-26 21:01:29 +00:00 by bsakaguchi · 2 comments
Owner

Originally created by @intervitens on GitHub (Jan 11, 2024).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-MoE/issues/4

Error log: https://sprunge.us/MefNDs
Example I'm trying to run: http://sprunge.us/laQn8Q
Installed packages and versions: http://sprunge.us/ich59c
The only thing I changed from the example in the HF repo is the device map to prevent OOM on my dual GPU setup.

EDIT: seems to be some kind of an issue with my OS/hardware, cannot replicate on a rented machine.

Originally created by @intervitens on GitHub (Jan 11, 2024). Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-MoE/issues/4 Error log: https://sprunge.us/MefNDs Example I'm trying to run: http://sprunge.us/laQn8Q Installed packages and versions: http://sprunge.us/ich59c The only thing I changed from the example in the HF repo is the device map to prevent OOM on my dual GPU setup. EDIT: seems to be some kind of an issue with my OS/hardware, cannot replicate on a rented machine.
Author
Owner

@intervitens commented on GitHub (Jan 11, 2024):

Closing issue, since this doesn't seem to be related to the model itself.

<!-- gh-comment-id:1888045447 --> @intervitens commented on GitHub (Jan 11, 2024): Closing issue, since this doesn't seem to be related to the model itself.
Author
Owner

@intervitens commented on GitHub (Jan 14, 2024):

Downgrading the NVIDIA driver from version 545 to 535 solved the issue.

<!-- gh-comment-id:1890922424 --> @intervitens commented on GitHub (Jan 14, 2024): Downgrading the NVIDIA driver from version 545 to 535 solved the issue.
Sign in to join this conversation.
No labels
pull-request
WIP
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepSeek-MoE#3
No description provided.