Skip to content

[FixBug]online serving fails for high-resolution videos #198

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Gaohan123 merged 3 commits into from
Dec 5, 2025
Merged

[FixBug]online serving fails for high-resolution videos #198

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fe2bb7f
fix online serving fails for high-resolution videos
princepride Dec 4, 2025
5fa2b49
revert yaml
princepride Dec 4, 2025
e60b8e5
Merge branch 'main' into fix-qwen3-omni-high-resolution
Gaohan123 Dec 5, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions vllm_omni/model_executor/models/qwen3_omni/qwen3_omni_moe_thinker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1415,7 +1415,7 @@ def forward(
inputs_embeds = None

if self.use_deepstack and inputs_embeds is not None and get_pp_group().is_first_rank:
deepstack_input_embeds = self._get_deepstack_input_embeds(inputs_embeds.size(0))
deepstack_input_embeds = self._get_deepstack_input_embeds(input_ids.size(1))
else:
deepstack_input_embeds = None

Expand All @@ -1431,7 +1431,7 @@ def forward(
)

if inputs_embeds is not None and get_pp_group().is_first_rank:
self._clear_deepstack_input_embeds(inputs_embeds.size(0))
self._clear_deepstack_input_embeds(input_ids.size(1))

return hidden_states.unsqueeze(0), captured_hidden_states

Expand Down
4 changes: 2 additions & 2 deletions vllm_omni/model_executor/stage_configs/qwen3_omni_moe.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ stage_args:
model_arch: Qwen3OmniMoeForConditionalGeneration
worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler
gpu_memory_utilization: 0.6
gpu_memory_utilization: 0.5
enforce_eager: true
trust_remote_code: true
engine_output_type: latent # Output hidden states for talker
Expand Down Expand Up @@ -43,7 +43,7 @@ stage_args:
model_arch: Qwen3OmniMoeForConditionalGeneration
worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler
gpu_memory_utilization: 0.3
gpu_memory_utilization: 0.45
enforce_eager: true
trust_remote_code: true
engine_output_type: latent # Output codec codes for code2wav
Expand Down