[Edge][WP1] Add edge deploy profile and deploy_profile selector

* vllm_omni/deploy/edge/qwen3_tts.yaml: single-device Qwen3-TTS profile with a
  ramped codec chunk schedule (2->4->8->16->25 frames) that removes the ~80 ms
  playback stall of the default 1->25 schedule, decode batch 1, max_num_seqs 4.
* deploy_profile="edge" (Omni/AsyncOmni kwarg, --deploy-profile CLI, headless
  serve) resolves vllm_omni/deploy/<profile>/<model_type>.yaml; mutually
  exclusive with deploy_config; explicit profiles never fall back silently.
* tests/helpers/stage_config.py: modify_stage_config anchors a relative
  base_config before writing the temp YAML so overlay-derived configs resolve.
* async_omni_engine.py also carries the init-timeline hooks used by WP2.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
tzhouamandClaude Fable 5.1 committed 2026-09-09 10:27:33 +08:00
1 parent b70207f57f
commit b22dcc59f7
7 files changed
+278

No files matched your search

+3
View File
@@ -518,6 +518,9 @@ class OrchestratorArgs:
# === Config Files ===
deploy_config: str | None = None
# Named deploy profile (e.g. ``edge``) resolved to ``deploy/<profile>/<model_type>.yaml``
# by the orchestrator; mutually exclusive with ``deploy_config``.
deploy_profile: str | None = None
stage_overrides: str | None = None # raw JSON string; parsed downstream
# Optional composable-parallel strategy.yaml; orchestrator reads it, overlays
# derived sizing onto merged stages, then drops it before per-stage engine args.