0 Fork 0

Files

T

hsliuustc0106 9a7ea9b003 [Bugfix] Reconcile env-var inventory with post-#6217 main drift

Merge-window drift broke the new environment-variable ratchet on main
(build 14119): #6255 added the VLLM_OMNI_ASYNC_OUTPUT_TIMEOUT read in
the diffusion engine and #6362 removed the last package reference to
ROSVOT_SOURCE_DIR, both after #6217's CI run.

- Classify VLLM_OMNI_ASYNC_OUTPUT_TIMEOUT as PUBLIC_OMNI and document
  it in the environment-variables reference page.
- Drop the stale ROSVOT_SOURCE_DIR model-specific row now that the
  SoulX-Singer implementation is gone.
- Update the reviewed snapshot counts and disposition tables to match.

Signed-off-by: hsliuustc0106 <liuhongsheng4@huawei.com>

2026-08-26 06:06:04 +08:00

composable_parallel.md

[Docs] Remove stale stage config references (#6270 )

2026-08-20 07:09:12 +08:00

environment_variables.md

[Bugfix] Reconcile env-var inventory with post-#6217 main drift

2026-08-26 06:06:04 +08:00

gpu_memory_utilization.md

[Docs] Remove stale stage config references (#6270 )

2026-08-20 07:09:12 +08:00

README.md

[Doc] Add environment variables configuration reference (#6217 )

2026-08-26 04:58:54 +08:00

stage_configs.md

[Core] Add UniProc diffusion executor for single-GPU (#6308 )

2026-08-22 21:58:21 +08:00

README.md

Configuration Options

This section lists the most common options for running vLLM-Omni.

For options within a vLLM Engine, please refer to the vLLM 0.26 configuration guide.

Each model defines fixed topology in a registered PipelineConfig and runtime overrides in a deploy YAML.

For process-level settings shared by the CLI, deploy YAML, workers, and model integrations, see Environment Variables.

For a specific example, see the Qwen2.5-Omni deploy config. The matching frozen pipeline topology lives at vllm_omni/model_executor/models/qwen2_5_omni/pipeline.py.

For an introduction, see Pipeline and deploy configurations.

Memory Configuration

GPU Memory Calculation and Configuration - Guide on how to calculate memory requirements and set up gpu_memory_utilization for optimal performance

Optimization Features

Diffusion Features Overview - Complete overview of all diffusion model features and supported models