[Config Refactor][2/N] Pipeline + Deploy Config Schema (#2383)

Signed-off-by: lishunyang <lishunyang12@163.com>
Signed-off-by: reidliu41 <reid201711@gmail.com>
Signed-off-by: Alex Brooks <albrooks@redhat.com>
Co-authored-by: reidliu41 <reid201711@gmail.com>
Co-authored-by: xiaohajiayou <75477391+xiaohajiayou@users.noreply.github.com>
Co-authored-by: Alex Brooks <albrooks@redhat.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
authored and GitHub committed 2026-04-19 14:36:00 +12:00
1 parent a683b1dd75
commit a3903810fe
104 files changed
+3959 -3814

No files matched your search

@@ -0,0 +1,83 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Qwen2.5-Omni pipeline topology (frozen).
Stage 0: Thinker — multimodal understanding + text generation
Stage 1: Talker — text embeddings → speech tokens
Stage 2: Code2Wav — speech tokens → audio waveform
"""
from vllm_omni.config.stage_config import (
PipelineConfig,
StageExecutionType,
StagePipelineConfig,
register_pipeline,
)
_PROC = "vllm_omni.model_executor.stage_input_processors.qwen2_5_omni"
QWEN2_5_OMNI_PIPELINE = PipelineConfig(
model_type="qwen2_5_omni",
model_arch="Qwen2_5OmniForConditionalGeneration",
stages=(
StagePipelineConfig(
stage_id=0,
model_stage="thinker",
execution_type=StageExecutionType.LLM_AR,
input_sources=(),
final_output=True,
final_output_type="text",
owns_tokenizer=True,
requires_multimodal_data=True,
engine_output_type="latent",
sampling_constraints={"detokenize": True},
),
StagePipelineConfig(
stage_id=1,
model_stage="talker",
execution_type=StageExecutionType.LLM_AR,
input_sources=(0,),
engine_output_type="latent",
custom_process_input_func=f"{_PROC}.thinker2talker",
sampling_constraints={
"detokenize": True,
"stop_token_ids": [8294],
},
),
StagePipelineConfig(
stage_id=2,
model_stage="code2wav",
execution_type=StageExecutionType.LLM_GENERATION,
input_sources=(1,),
final_output=True,
final_output_type="audio",
engine_output_type="audio",
sampling_constraints={"detokenize": True},
),
),
)
register_pipeline(QWEN2_5_OMNI_PIPELINE)
# Single-stage thinker-only variant for the abort test.
QWEN2_5_OMNI_THINKER_ONLY_PIPELINE = PipelineConfig(
model_type="qwen2_5_omni_thinker_only",
model_arch="Qwen2_5OmniForConditionalGeneration",
stages=(
StagePipelineConfig(
stage_id=0,
model_stage="thinker",
execution_type=StageExecutionType.LLM_AR,
input_sources=(),
final_output=True,
final_output_type="text",
owns_tokenizer=True,
requires_multimodal_data=True,
engine_output_type="latent",
sampling_constraints={"detokenize": True},
),
),
)
register_pipeline(QWEN2_5_OMNI_THINKER_ONLY_PIPELINE)
@@ -0,0 +1,66 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Qwen3-Omni-MoE pipeline topology (frozen).
Stage 0: Thinker — multimodal understanding + text generation
Stage 1: Talker — text embeddings → RVQ codec codes
Stage 2: Code2Wav — RVQ codes → audio waveform
"""
from vllm_omni.config.stage_config import (
PipelineConfig,
StageExecutionType,
StagePipelineConfig,
register_pipeline,
)
_PROC = "vllm_omni.model_executor.stage_input_processors.qwen3_omni"
QWEN3_OMNI_PIPELINE = PipelineConfig(
model_type="qwen3_omni_moe",
model_arch="Qwen3OmniMoeForConditionalGeneration",
stages=(
StagePipelineConfig(
stage_id=0,
model_stage="thinker",
execution_type=StageExecutionType.LLM_AR,
input_sources=(),
final_output=True,
final_output_type="text",
owns_tokenizer=True,
requires_multimodal_data=True,
hf_config_name="thinker_config",
engine_output_type="latent",
custom_process_next_stage_input_func=(f"{_PROC}.thinker2talker_async_chunk"),
sampling_constraints={"detokenize": True},
),
StagePipelineConfig(
stage_id=1,
model_stage="talker",
execution_type=StageExecutionType.LLM_AR,
input_sources=(0,),
hf_config_name="talker_config",
engine_output_type="latent",
custom_process_input_func=f"{_PROC}.thinker2talker",
custom_process_next_stage_input_func=(f"{_PROC}.talker2code2wav_async_chunk"),
sampling_constraints={
"detokenize": False,
"stop_token_ids": [2150],
},
),
StagePipelineConfig(
stage_id=2,
model_stage="code2wav",
execution_type=StageExecutionType.LLM_GENERATION,
input_sources=(1,),
final_output=True,
final_output_type="audio",
hf_config_name="thinker_config",
engine_output_type="audio",
custom_process_input_func=f"{_PROC}.talker2code2wav",
sampling_constraints={"detokenize": True},
),
),
)
register_pipeline(QWEN3_OMNI_PIPELINE)
@@ -0,0 +1,51 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Qwen3-TTS pipeline: Talker (text → RVQ codec) → Code2Wav (codec → audio).
Chunked vs end-to-end mode is dispatched from ``deploy.async_chunk``.
"""
from vllm_omni.config.stage_config import (
PipelineConfig,
StageExecutionType,
StagePipelineConfig,
register_pipeline,
)
_PROC = "vllm_omni.model_executor.stage_input_processors.qwen3_tts"
QWEN3_TTS_PIPELINE = PipelineConfig(
model_type="qwen3_tts",
# Pipeline-level default; the code2wav stage overrides per-stage below.
model_arch="Qwen3TTSTalkerForConditionalGeneration",
stages=(
StagePipelineConfig(
stage_id=0,
model_stage="qwen3_tts",
execution_type=StageExecutionType.LLM_AR,
input_sources=(),
owns_tokenizer=True,
engine_output_type="latent",
async_chunk_process_next_stage_input_func=(f"{_PROC}.talker2code2wav_async_chunk"),
sampling_constraints={
"detokenize": False,
"stop_token_ids": [2150],
},
),
StagePipelineConfig(
stage_id=1,
model_stage="code2wav",
execution_type=StageExecutionType.LLM_GENERATION,
input_sources=(0,),
final_output=True,
final_output_type="audio",
engine_output_type="audio",
model_arch="Qwen3TTSCode2Wav",
sync_process_input_func=f"{_PROC}.talker2code2wav",
sampling_constraints={"detokenize": True},
extras={"tts_args": {"max_instructions_length": 500}},
),
),
)
register_pipeline(QWEN3_TTS_PIPELINE)
@@ -1,92 +0,0 @@
model_type: qwen3_tts
async_chunk: true
stages:
- stage_id: 0
model_stage: qwen3_tts
stage_type: llm
is_comprehension: true
input_sources: []
worker_type: ar
scheduler_cls: vllm_omni.core.sched.omni_ar_scheduler.OmniARScheduler
runtime:
devices: "0"
engine_args:
max_num_seqs: 10
model_arch: Qwen3TTSTalkerForConditionalGeneration
hf_overrides:
architectures: [Qwen3TTSTalkerForConditionalGeneration]
enforce_eager: false
trust_remote_code: true
async_scheduling: true
enable_prefix_caching: false
engine_output_type: latent
gpu_memory_utilization: 0.08
distributed_executor_backend: "mp"
max_num_batched_tokens: 512
max_model_len: 4096
custom_process_next_stage_input_func: vllm_omni.model_executor.stage_input_processors.qwen3_tts.talker2code2wav_async_chunk
output_connectors:
to_stage_1: connector_of_shared_memory
default_sampling_params:
temperature: 0.9
top_k: 50
max_tokens: 4096
seed: 42
detokenize: false
repetition_penalty: 1.05
stop_token_ids: [2150]
- stage_id: 1
model_stage: code2wav
stage_type: llm
input_sources: [0]
worker_type: generation
scheduler_cls: vllm_omni.core.sched.omni_generation_scheduler.OmniGenerationScheduler
final_output: true
final_output_type: audio
runtime:
devices: "0"
engine_args:
max_num_seqs: 1
model_arch: Qwen3TTSCode2Wav
hf_overrides:
architectures: [Qwen3TTSCode2Wav]
enforce_eager: true
trust_remote_code: true
async_scheduling: true
enable_prefix_caching: false
engine_output_type: audio
gpu_memory_utilization: 0.08
distributed_executor_backend: "mp"
max_num_batched_tokens: 65536
max_model_len: 65536
input_connectors:
from_stage_0: connector_of_shared_memory
tts_args:
max_instructions_length: 500
default_sampling_params:
temperature: 0.0
top_p: 1.0
top_k: -1
max_tokens: 65536
seed: 42
detokenize: true
repetition_penalty: 1.0
connectors:
connector_of_shared_memory:
name: SharedMemoryConnector
extra:
shm_threshold_bytes: 65536
codec_streaming: true
connector_get_sleep_s: 0.01
connector_get_max_wait_first_chunk: 3000
connector_get_max_wait: 300
codec_chunk_frames: 25
# Match the decoder sliding attention window to avoid chunk-boundary noise.
codec_left_context_frames: 72
edges:
- from: 0
to: 1