26 lines
601 B
YAML
26 lines
601 B
YAML
# Explicit CUDA Model Runner V2 profile. `qwen3_tts.yaml` now selects V2 by
|
|
# default (experimental); this file is kept for scripts that name the MRV2
|
|
# profile explicitly and for the platform fallbacks below.
|
|
base_config: qwen3_tts.yaml
|
|
model_runner: v2
|
|
|
|
platforms:
|
|
npu:
|
|
model_runner: v1
|
|
xpu:
|
|
model_runner: v1
|
|
rocm:
|
|
model_runner: v1
|
|
musa:
|
|
model_runner: v1
|
|
|
|
stages:
|
|
- stage_id: 0
|
|
# Bound per-step prefill work, matching the high-concurrency TTS profile.
|
|
max_num_batched_tokens: 512
|
|
|
|
connectors:
|
|
connector_of_shared_memory:
|
|
extra:
|
|
talker_first_audio: true
|