hailin
/
taixf
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

feat: enable flash_attention_2 for Qwen3-TTS

This commit is contained in:
hailin 2026-04-07 03:22:48 -07:00
parent 78bc3f71c0
commit b75e813c03
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
modules/tts/qwen3_tts.py
View File

@ -41,6 +41,7 @@ class TTSProvider(TTSProviderBase):
model_path, model_path,
device_map=device, device_map=device,
dtype=dtype, dtype=dtype,
attn_implementation="flash_attention_2",
) )
self.tokenizer = Qwen3TTSTokenizer.from_pretrained(tokenizer_path) self.tokenizer = Qwen3TTSTokenizer.from_pretrained(tokenizer_path)