[Bugfix] Fix Ultravox on V1 (#14929)

Signed-off-by: DarkLight1337 <[email protected]>

Author: DarkLight1337

vllm/model_executor/models/ultravox.py

@@ -5,7 +5,7 @@
 import math
 from collections.abc import Iterable, Mapping, Sequence
 from functools import cached_property
-from typing import Any, List, Literal, Optional, Set, Tuple, TypedDict, Union
+from typing import Any, Literal, Optional, Set, Tuple, TypedDict, Union
 
 import torch
 import torch.utils.checkpoint
@@ -36,7 +36,7 @@
 from vllm.transformers_utils.configs.ultravox import UltravoxConfig
 
 from .interfaces import (MultiModalEmbeddings, SupportsLoRA,
-                         SupportsMultiModal, SupportsPP, SupportsV0Only)
+                         SupportsMultiModal, SupportsPP)
 from .utils import (AutoWeightsLoader, WeightsMapper, flatten_bn,
                     init_vllm_registered_model, maybe_prefix,
                     merge_multimodal_embeddings,
@@ -50,14 +50,14 @@
 
 class UltravoxAudioFeatureInputs(TypedDict):
     type: Literal["audio_features"]
-    data: NestedTensors
+    data: Union[torch.Tensor, list[torch.Tensor], list[list[torch.Tensor]]]
     """Shape: `(batch_size, num_chunks, 80, M)`"""
-    lens: NestedTensors
+    lens: Union[torch.Tensor, list[torch.Tensor]]
     """
     Length of the audio frames. Used for attention mask in WhisperEncoder.
     Shape: `(batch_size, num_chunks)`
     """
-    token_len: NestedTensors
+    token_len: Union[torch.Tensor, list[torch.Tensor]]
     """
     Length of the audio tokens. Used for flattening the audio features.
     Shape: `(batch_size, num_chunks)`
@@ -405,8 +405,7 @@ def forward(
     UltravoxMultiModalProcessor,
     info=UltravoxProcessingInfo,
     dummy_inputs=UltravoxDummyInputsBuilder)
-class UltravoxModel(nn.Module, SupportsMultiModal, SupportsPP, SupportsLoRA,
-                    SupportsV0Only):
+class UltravoxModel(nn.Module, SupportsMultiModal, SupportsPP, SupportsLoRA):
 
     packed_modules_mapping = {
         "qkv_proj": ["q_proj", "k_proj", "v_proj"],
@@ -506,6 +505,12 @@ def _parse_and_validate_audio_input(
             if not isinstance(audio_features, (torch.Tensor, list)):
                 raise ValueError("Incorrect type of audio features. "
                                  f"Got type: {type(audio_features)}")
+            if not isinstance(audio_lens, (torch.Tensor, list)):
+                raise ValueError("Incorrect type of audio_lens. "
+                                 f"Got type: {type(audio_features)}")
+            if not isinstance(audio_token_len, (torch.Tensor, list)):
+                raise ValueError("Incorrect type of audio_token_len. "
+                                 f"Got type: {type(audio_features)}")
 
             return UltravoxAudioFeatureInputs(type="audio_features",
                                               data=audio_features,
@@ -523,21 +528,19 @@ def _parse_and_validate_audio_input(
         raise AssertionError("This line should be unreachable.")
 
     def _process_audio_input(
-            self, audio_input: UltravoxAudioInputs) -> NestedTensors:
+        self,
+        audio_input: UltravoxAudioInputs,
+    ) -> Union[NestedTensors, tuple[torch.Tensor, ...]]:
         if audio_input["type"] == "audio_embeds":
             return audio_input["data"]
 
         # Pad and concatenate audio features
         # [[B1, 80, M1], [B2, 80, M2]] -> [B1+B2, 80, max(M1, M2)]
         audio_features = pad_and_concat_to_dim3(audio_input["data"])
 
-        if isinstance(audio_input['lens'], list):
-            # [B1, B2] -> [B1+B2]
-            audio_lens = torch.cat(audio_input['lens'])
-            audio_token_len = torch.cat(audio_input['token_len'])
-        else:
-            audio_lens = flatten_bn(audio_input['lens'])
-            audio_token_len = flatten_bn(audio_input['token_len'])
+        # [B1, B2] -> [B1+B2]
+        audio_lens = flatten_bn(audio_input['lens'], concat=True)
+        audio_token_len = flatten_bn(audio_input['token_len'], concat=True)
 
         embeddings = self._audio_features_to_embeddings(
             audio_features, audio_lens)
@@ -554,7 +557,12 @@ def _process_audio_input(
         # Apply mask and flatten
         flattened_embeddings = embeddings[mask]
 
-        return flattened_embeddings
+        # Return one tensor per input audio
+        embed_lens = [
+            token_len_item.sum().item()
+            for token_len_item in audio_input['token_len']
+        ]
+        return flattened_embeddings.split(embed_lens)
 
     def get_multimodal_embeddings(
             self, **kwargs: object) -> Optional[MultiModalEmbeddings]:
@@ -646,7 +654,7 @@ def load_weights(self, weights: Iterable[Tuple[str,
 
 
 def pad_and_concat_to_dim3(
-    features: Union[torch.Tensor, List[torch.Tensor], List[List[torch.Tensor]]]
+    features: Union[torch.Tensor, list[torch.Tensor], list[list[torch.Tensor]]]
 ) -> torch.Tensor:
     """
     Pad and concatenate a list of tensors.