mirror of
https://github.com/deepseek-ai/DeepSeek-R1.git
synced 2026-11-05 00:47:42 +00:00
[GH-ISSUE #144] Optimized for Vulcan (AMD GPU) #118
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
deepseek-ai/DeepSeek-R1#118
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @BeMgCaSrBa on GitHub (Jan 29, 2025).
Original GitHub issue: https://github.com/deepseek-ai/DeepSeek-R1/issues/144
I encountered some problems when using Vulcan. On the one hand, the reasoning speed was slow; on the other hand, some models could not occupy all the video memory, but started to use the shared video memory when the video memory was not full (especially when I was using the model "Deep Seek-R1-Distill-QWEN-32b-Q4 _ K _ M"). Can more optimization be made for Vulcan?
@zekosan commented on GitHub (Jan 30, 2025):
Im having the same issue. but im using Rx 6600 with deepseek-r1:1.5b model. Its only using CPU and RAM and not utilizing the GPU or VRAM at all. Can anyone help?
@AlanGreyjoy commented on GitHub (Jan 30, 2025):
what is wrong with both of you....
@dinithaw commented on GitHub (Feb 1, 2025):
Vulcan? Don't be stupid. Vulcan is graphic rendering API. If u want support for AMD gpus its ML api called ROCm or something. It's like dumber version of nvidia cuda. Using AMD for ML works are dumb anyway 😂
@github-actions[bot] commented on GitHub (Mar 12, 2025):
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. If you believe this issue is still relevant, please leave a comment to keep it open. Thank you for your contributions!
@github-actions[bot] commented on GitHub (Apr 7, 2025):
false