From 61e170c69ab1e5f386c110f734779fdd50914301 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E6=B1=AA=E5=BF=97=E9=B9=8F?= Date: Wed, 18 Mar 2026 19:46:44 +0800 Subject: [PATCH] [Feature]: Remove some useless `hf_overrides` in yaml (#1898) Signed-off-by: princepride --- benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs1.yaml | 4 ---- benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs4.yaml | 4 ---- tests/perf/stage_configs/qwen3_tts.yaml | 4 ---- .../model_executor/stage_configs/fish_speech_s2_pro.yaml | 4 ---- vllm_omni/model_executor/stage_configs/qwen3_tts.yaml | 6 ------ vllm_omni/model_executor/stage_configs/qwen3_tts_batch.yaml | 6 ------ .../stage_configs/qwen3_tts_no_async_chunk.yaml | 4 ---- vllm_omni/platforms/npu/stage_configs/qwen3_tts.yaml | 6 ------ 8 files changed, 38 deletions(-) diff --git a/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs1.yaml b/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs1.yaml index 1597f8aa2..d8e4072c0 100644 --- a/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs1.yaml +++ b/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs1.yaml @@ -10,8 +10,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -43,8 +41,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs4.yaml b/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs4.yaml index 18493f3ae..93692d60d 100644 --- a/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs4.yaml +++ b/benchmarks/qwen3-tts/vllm_omni/configs/qwen3_tts_bs4.yaml @@ -11,8 +11,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -44,8 +42,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/tests/perf/stage_configs/qwen3_tts.yaml b/tests/perf/stage_configs/qwen3_tts.yaml index 5de7f13f1..1ad568f26 100644 --- a/tests/perf/stage_configs/qwen3_tts.yaml +++ b/tests/perf/stage_configs/qwen3_tts.yaml @@ -14,8 +14,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -47,8 +45,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/vllm_omni/model_executor/stage_configs/fish_speech_s2_pro.yaml b/vllm_omni/model_executor/stage_configs/fish_speech_s2_pro.yaml index 7f5ba1202..a847cb181 100644 --- a/vllm_omni/model_executor/stage_configs/fish_speech_s2_pro.yaml +++ b/vllm_omni/model_executor/stage_configs/fish_speech_s2_pro.yaml @@ -8,8 +8,6 @@ stage_args: engine_args: model_stage: fish_speech_slow_ar model_arch: FishSpeechSlowARForConditionalGeneration - hf_overrides: - architectures: [FishSpeechSlowARForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -43,8 +41,6 @@ stage_args: engine_args: model_stage: dac_decoder model_arch: FishSpeechDACDecoder - hf_overrides: - architectures: [FishSpeechDACDecoder] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/vllm_omni/model_executor/stage_configs/qwen3_tts.yaml b/vllm_omni/model_executor/stage_configs/qwen3_tts.yaml index 971c0de12..f92874bf6 100644 --- a/vllm_omni/model_executor/stage_configs/qwen3_tts.yaml +++ b/vllm_omni/model_executor/stage_configs/qwen3_tts.yaml @@ -9,9 +9,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -44,9 +41,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/vllm_omni/model_executor/stage_configs/qwen3_tts_batch.yaml b/vllm_omni/model_executor/stage_configs/qwen3_tts_batch.yaml index 7770b9e7c..23557bf23 100644 --- a/vllm_omni/model_executor/stage_configs/qwen3_tts_batch.yaml +++ b/vllm_omni/model_executor/stage_configs/qwen3_tts_batch.yaml @@ -13,9 +13,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -48,9 +45,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/vllm_omni/model_executor/stage_configs/qwen3_tts_no_async_chunk.yaml b/vllm_omni/model_executor/stage_configs/qwen3_tts_no_async_chunk.yaml index b4ae91c9a..f559bdebd 100644 --- a/vllm_omni/model_executor/stage_configs/qwen3_tts_no_async_chunk.yaml +++ b/vllm_omni/model_executor/stage_configs/qwen3_tts_no_async_chunk.yaml @@ -9,8 +9,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: false @@ -39,8 +37,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true diff --git a/vllm_omni/platforms/npu/stage_configs/qwen3_tts.yaml b/vllm_omni/platforms/npu/stage_configs/qwen3_tts.yaml index 53185e453..530e957f0 100644 --- a/vllm_omni/platforms/npu/stage_configs/qwen3_tts.yaml +++ b/vllm_omni/platforms/npu/stage_configs/qwen3_tts.yaml @@ -9,9 +9,6 @@ stage_args: engine_args: model_stage: qwen3_tts model_arch: Qwen3TTSTalkerForConditionalGeneration - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSTalkerForConditionalGeneration] worker_type: ar scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler enforce_eager: true @@ -44,9 +41,6 @@ stage_args: engine_args: model_stage: code2wav model_arch: Qwen3TTSCode2Wav - # Force stage-specific registered architecture. - hf_overrides: - architectures: [Qwen3TTSCode2Wav] worker_type: generation scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler enforce_eager: true