[Feature] Support flexible task_type configuration for Qwen3-TTS models (#1197)

Signed-off-by: Jack Lee <695697442@qq.com>
Signed-off-by: jackleehal <695697442@qq.com>
Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com>
Co-authored-by: JackLeeHal <15664755+JackLeeHal@users.noreply.github.com>
This commit is contained in:
authored and GitHub committed 2026-03-05 13:39:09 -05:00
1 parent e643aaec15
commit c0b714a8ac
4 files changed
+22 -1

No files matched your search

@@ -94,7 +94,9 @@ class Qwen3TTSModelForGeneration(nn.Module):
torch_dtype=torch.bfloat16,
**attn_kwargs,
)
self.task_type = _normalize_task_type(model_path.split("-")[-1].split("/")[0])
self.task_type = getattr(vllm_config.model_config, "task_type", None) or _normalize_task_type(
model_path.split("-")[-1].split("/")[0]
)
# Mark that this model produces multimodal outputs
self.have_multimodal_outputs = True