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

[Model] SupportHunyuanImage3 Diffusion Model in vllm-omni (#1085)

Browse Source 
Signed-off-by: Semmer2 <semmer@live.cn>
This commit is contained in:
Jiaping Wu authored and GitHub committed 2026-02-10 03:03:44 +13:00
1 parent 26ba1e4de8
commit 5fea4823b7
13 files changed
+5695 -23
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

vllm_omni/diffusion/worker/diffusion_model_runner.py
+10 -11
View File
@@ -87,18 +87,17 @@ class DiffusionModelRunner:
return nullcontext()
# Load model within forward context
with set_forward_context(vllm_config=self.vllm_config, omni_diffusion_config=self.od_config):
load_config = LoadConfig()
model_loader = DiffusersPipelineLoader(load_config)
time_before_load = time.perf_counter()
load_config = LoadConfig()
model_loader = DiffusersPipelineLoader(load_config)
time_before_load = time.perf_counter()
with get_memory_context():
with DeviceMemoryProfiler() as m:
self.pipeline = model_loader.load_model(
od_config=self.od_config,
load_device=load_device,
)
time_after_load = time.perf_counter()
with get_memory_context():
with DeviceMemoryProfiler() as m:
self.pipeline = model_loader.load_model(
od_config=self.od_config,
load_device=load_device,
)
time_after_load = time.perf_counter()
logger.info(
"Model loading took %.4f GiB and %.6f seconds",