anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

[Misc] fix yaml config (#320)

Browse Source 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
This commit is contained in:
Lancer authored and GitHub committed 2025-12-16 02:40:30 +00:00
1 parent 06c2b7f6e4
commit 31fedcfa2b
5 files changed
+13 -13
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

vllm_omni/model_executor/stage_configs/qwen3_omni_moe_multiconnector.yaml
+3 -3
View File
@@ -13,7 +13,7 @@ stage_args:
model_stage: thinker
model_arch: Qwen3OmniMoeForConditionalGeneration
worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler
scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
gpu_memory_utilization: 0.6
enforce_eager: true
trust_remote_code: true
@@ -45,7 +45,7 @@ stage_args:
model_stage: talker
model_arch: Qwen3OmniMoeForConditionalGeneration
worker_cls: vllm_omni.worker.gpu_ar_worker.GPUARWorker
scheduler_cls: vllm_omni.core.sched.scheduler.OmniScheduler
scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
gpu_memory_utilization: 0.3
enforce_eager: true
trust_remote_code: true
@@ -80,7 +80,7 @@ stage_args:
model_stage: code2wav
model_arch: Qwen3OmniMoeForConditionalGeneration
worker_cls: vllm_omni.worker.gpu_generation_worker.GPUGenerationWorker
scheduler_cls: vllm_omni.core.sched.generation_scheduler.GenerationScheduler
scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler
enforce_eager: true
trust_remote_code: true
enable_prefix_caching: false