Skip to content

[GPU] fix dGPU performance issue when run qwen3_moe models#35135

Merged
e-ddykim merged 2 commits intoopenvinotoolkit:masterfrom
riverlijunjie:river/fixed_dGPU_perf_issue
Apr 3, 2026
Merged

[GPU] fix dGPU performance issue when run qwen3_moe models#35135
e-ddykim merged 2 commits intoopenvinotoolkit:masterfrom
riverlijunjie:river/fixed_dGPU_perf_issue

Conversation

@riverlijunjie
Copy link
Copy Markdown
Contributor

@riverlijunjie riverlijunjie commented Apr 3, 2026

Details:

  • Fix dGPU performance issue when run qwen3 moe models
  • It should allocate usm_device if the buffer doesn't copy_to/from cpu memory, especially for dGPU.

Tickets:

AI Assistance:

  • AI assistance used: no / yes
  • If yes, summarize how AI was used and what human validation was performed (build/tests/manual checks).

@riverlijunjie riverlijunjie requested review from a team as code owners April 3, 2026 02:02
@github-actions github-actions bot added the category: GPU OpenVINO GPU plugin label Apr 3, 2026
@riverlijunjie riverlijunjie changed the base branch from river/fixed_dGPU_perf_issue to master April 3, 2026 02:05
Copy link
Copy Markdown
Contributor

@e-ddykim e-ddykim left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me

@e-ddykim e-ddykim enabled auto-merge April 3, 2026 02:39
@e-ddykim e-ddykim added this pull request to the merge queue Apr 3, 2026
Merged via the queue into openvinotoolkit:master with commit 5e09f61 Apr 3, 2026
188 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

category: GPU OpenVINO GPU plugin

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants