Isolate Qwen-Image 2.1 prefill and decode scheduling (#7770)

Signed-off-by: Tianyu Guo <guoty9@mail2.sysu.edu.cn>
This commit is contained in:
Tianyu Guo authored and GitHub committed 2026-09-18 14:11:22 +08:00
1 parent a26013145c
commit b0a309e5e7
4 files changed
+27 -3

No files matched your search

+2
View File
@@ -73,6 +73,8 @@ class OmniDiffusionRequest:
# A model preprocessor may keep selected requests on the legacy full-forward
# path even when the engine is globally configured for step execution.
use_step_execution: bool = True
# Whether first and later denoising steps may share a step batch.
allow_mixed_step_phases: bool = True
# KV-recv wall-clock (ms), set by the runner's _prepare_request_for_forward
# and carried to DiffusionOutput for the vllm_omni:diffusion_kv_load_s metric.
kv_recv_ms: float = 0.0