[Model] Consolidate pooler implementations (#20927)

DarkLight1337 · web-flow · commit 1c3198b6c4e5 · 2025-07-16T13:39:13.000Z
Signed-off-by: DarkLight1337 &lt;tlleungac@connect.ust.hk&gt;
diff --git a/vllm/model_executor/layers/pooler.py b/vllm/model_executor/layers/pooler.py
diff --git a/vllm/model_executor/models/adapters.py b/vllm/model_executor/models/adapters.py
@@ -58,22 +58,27 @@ def __init__(
         ) -> None:
             super().__init__(vllm_config=vllm_config, prefix=prefix, **kwargs)
 
+            self.vllm_config = vllm_config
+
             # These are not used in pooling models
             for attr in ("lm_head", "logits_processor"):
                 if hasattr(self, attr):
                     delattr(self, attr)
 
+            # If the model already defines a pooler instance, don't overwrite it
+            if not getattr(self, "_pooler", None):
+                self._init_pooler(vllm_config, prefix=prefix)
+
+        def _init_pooler(self, vllm_config: "VllmConfig", prefix: str = ""):
             pooler_config = vllm_config.model_config.pooler_config
             assert pooler_config is not None
 
-            # If the model already defines a pooler instance, don't overwrite it
-            if not getattr(self, "_pooler", None):
-                self._pooler = Pooler.from_config_with_defaults(
-                    pooler_config,
-                    pooling_type=default_pooling_type,
-                    normalize=default_normalize,
-                    softmax=default_softmax,
-                )
+            self._pooler = Pooler.from_config_with_defaults(
+                pooler_config,
+                pooling_type=default_pooling_type,
+                normalize=default_normalize,
+                softmax=default_softmax,
+            )
 
         def pooler(
             self,
@@ -165,7 +170,9 @@ def as_seq_cls_model(cls: _T) -> _T:
 
     # Lazy import
     from vllm.model_executor.layers.linear import RowParallelLinear
-    from vllm.model_executor.layers.pooler import PoolerOutput, PoolingType
+    from vllm.model_executor.layers.pooler import (ClassifierPooler,
+                                                   PoolerOutput, PoolingType,
+                                                   SimplePooler)
     from vllm.model_executor.models.interfaces import SupportsCrossEncoding
     from vllm.model_executor.pooling_metadata import PoolingMetadata
     from vllm.sequence import IntermediateTensors
@@ -182,30 +189,40 @@ def as_seq_cls_model(cls: _T) -> _T:
     class ModelForSequenceClassification(ModelForPooling,
                                          SupportsCrossEncoding):
 
-        def __init__(
-            self,
-            *,
-            vllm_config: "VllmConfig",
-            prefix: str = "",
-            **kwargs: Any,
-        ) -> None:
-            super().__init__(vllm_config=vllm_config, prefix=prefix, **kwargs)
-
+        def _init_pooler(self, vllm_config: "VllmConfig", prefix: str = ""):
             config = vllm_config.model_config.hf_config
             quant_config = vllm_config.quant_config
 
-            self.vllm_config = vllm_config
-            self.task = vllm_config.model_config.task
-            self.pooling_type = (
-                vllm_config.model_config.pooler_config.pooling_type)
-
-            self.score = RowParallelLinear(config.hidden_size,
-                                           config.num_labels,
-                                           quant_config=quant_config,
-                                           input_is_parallel=False,
-                                           bias=False,
-                                           prefix=maybe_prefix(
-                                               prefix, "score"))
+            self.score = RowParallelLinear(
+                config.hidden_size,
+                config.num_labels,
+                input_is_parallel=False,
+                bias=False,
+                params_dtype=torch.float32,
+                quant_config=quant_config,
+                prefix=maybe_prefix(prefix, "score"),
+            )
+
+            pooler_config = vllm_config.model_config.pooler_config
+            assert pooler_config is not None
+
+            pooler = SimplePooler.from_config_with_defaults(
+                pooler_config,
+                pooling_type=PoolingType.LAST,
+                normalize=False,
+                softmax=True,
+            )
+
+            self._pooler = ClassifierPooler(
+                vllm_config.model_config,
+                pooling=pooler.pooling,
+                classifier=self._classifier,
+                act_fn=pooler.head.activation,
+            )
+
+        def _classifier(self, x: torch.Tensor):
+            x, _ = self.score(x.float())
+            return x
 
         def forward(
             self,
@@ -222,27 +239,7 @@ def pooler(
             hidden_states: Union[torch.Tensor, list[torch.Tensor]],
             pooling_metadata: PoolingMetadata,
         ) -> PoolerOutput:
-
-            def get_logits(hidden_states):
-                if isinstance(hidden_states, list):
-                    logits = [self.score(state)[0] for state in hidden_states]
-                else:
-                    logits, _ = self.score(hidden_states)
-                return logits
-
-            if self.pooling_type == PoolingType.ALL:
-                logits = get_logits(hidden_states)
-                return self._pooler(logits, pooling_metadata)
-            else:
-                hidden_states = self._pooler.extract_states(
-                    hidden_states, pooling_metadata)
-                logits = get_logits(hidden_states)
-                pooled_data = self._pooler.head(logits, pooling_metadata)
-
-                pooled_outputs = [
-                    self._pooler.build_output(data) for data in pooled_data
-                ]
-                return PoolerOutput(outputs=pooled_outputs)
+            return self._pooler(hidden_states, pooling_metadata)
 
         def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
             tokens = getattr(self.config, "classifier_from_token", None)
diff --git a/vllm/model_executor/models/bert.py b/vllm/model_executor/models/bert.py
@@ -2,7 +2,7 @@
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 
 from collections.abc import Iterable
-from typing import Optional
+from typing import Optional, Union
 
 import torch
 from torch import nn
@@ -18,7 +18,7 @@
                                                QKVParallelLinear,
                                                RowParallelLinear)
 from vllm.model_executor.layers.pooler import (ClassifierPooler, Pooler,
-                                               PoolingType)
+                                               PoolingMethod, PoolingType)
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding)
@@ -84,14 +84,18 @@ class BertPooler(nn.Module):
 
     def __init__(self, config: BertConfig):
         super().__init__()
+
+        self.pooling = PoolingMethod.from_pooling_type(PoolingType.CLS)
         self.dense = nn.Linear(config.hidden_size, config.hidden_size)
         self.activation = nn.Tanh()
 
-    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        # We "pool" the model by simply taking the hidden state corresponding
-        # to the first token.
-        first_token_tensor = hidden_states[0, :]
-        pooled_output = self.dense(first_token_tensor)
+    def forward(
+        self,
+        hidden_states: Union[torch.Tensor, list[torch.Tensor]],
+        pooling_metadata: PoolingMetadata,
+    ) -> Union[torch.Tensor, list[torch.Tensor]]:
+        pooled_output = self.pooling(hidden_states, pooling_metadata)
+        pooled_output = self.dense(pooled_output)
         pooled_output = self.activation(pooled_output)
         return pooled_output
 
@@ -472,8 +476,11 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
                               embedding_class=BertEmbedding,
                               add_pooling_layer=True)
         self.classifier = nn.Linear(config.hidden_size, config.num_labels)
-        self._pooler = ClassifierPooler(vllm_config.model_config,
-                                        self.classifier, self.bert.pooler)
+        self._pooler = ClassifierPooler(
+            vllm_config.model_config,
+            pooling=self.bert.pooler,
+            classifier=self.classifier,
+        )
 
     def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
         loader = AutoWeightsLoader(self)
diff --git a/vllm/model_executor/models/gritlm.py b/vllm/model_executor/models/gritlm.py
@@ -9,7 +9,7 @@
 
 from vllm.config import ModelConfig, VllmConfig
 from vllm.logger import init_logger
-from vllm.model_executor.layers.pooler import PoolerHead
+from vllm.model_executor.layers.pooler import PoolerHead, PoolerNormalize
 from vllm.model_executor.models.llama import LlamaForCausalLM
 from vllm.model_executor.pooling_metadata import (PoolingMetadata,
                                                   PoolingTensors)
@@ -49,7 +49,7 @@ def tokens_to_ids(tokens: list[str]) -> array:
         self.embed_pattern_ids = tokens_to_ids(
             ["▁<", "|", "embed", "|", ">", "<0x0A>"])
 
-        self.head = PoolerHead(normalize=True, softmax=False)
+        self.head = PoolerHead(PoolerNormalize())
 
     def _find_array(self, arr: array, target: array, start_idx: int) -> int:
         """
diff --git a/vllm/model_executor/models/interfaces.py b/vllm/model_executor/models/interfaces.py
@@ -659,7 +659,7 @@ def supports_cross_encoding(
 def has_step_pooler(model: Union[type[object], object]) -> bool:
     """Check if the model uses step pooler."""
     return is_pooling_model(model) and any(
-        type(module).__name__ == "StepPool" for module in model.modules())
+        type(module).__name__ == "StepPooler" for module in model.modules())
 
 
 class SupportsQuant:
diff --git a/vllm/model_executor/models/jamba.py b/vllm/model_executor/models/jamba.py
@@ -19,7 +19,8 @@
                                                RowParallelLinear)
 from vllm.model_executor.layers.logits_processor import LogitsProcessor
 from vllm.model_executor.layers.mamba.mamba_mixer import MambaMixer
-from vllm.model_executor.layers.pooler import Pooler, PoolingType
+from vllm.model_executor.layers.pooler import (ClassifierPooler, PoolingType,
+                                               SimplePooler)
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead, VocabParallelEmbedding)
@@ -564,29 +565,41 @@ class JambaForSequenceClassification(JambaForCausalLM):
 
     def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__(vllm_config=vllm_config, prefix=prefix)
+
         config = vllm_config.model_config.hf_config
         num_labels: int = config.num_labels
         score_bias: bool = getattr(config, 'score_bias', False)
-        self.score = nn.Linear(config.hidden_size, num_labels, bias=score_bias)
+
+        # TODO: The original reward weights have float32 accuracy data, we
+        # would like to load them in fp32 to get that extra precision.
+        # Currently weight_loader passes the weight which is already in bf16
+        self.score = nn.Linear(
+            config.hidden_size,
+            num_labels,
+            bias=score_bias,
+            dtype=torch.float32,
+        )
 
         pooler_config = vllm_config.model_config.pooler_config
-        self._pooler = Pooler.from_config_with_defaults(
+        assert pooler_config is not None
+
+        pooler = SimplePooler.from_config_with_defaults(
             pooler_config,
             pooling_type=PoolingType.LAST,
             normalize=False,
-            softmax=False)
+            softmax=False,
+        )
+
+        self._pooler = ClassifierPooler(
+            vllm_config.model_config,
+            pooling=pooler.pooling,
+            classifier=self.score,
+            act_fn=pooler.head.activation,
+        )
 
     def pooler(
         self,
         hidden_states: torch.Tensor,
         pooling_metadata: PoolingMetadata,
     ) -> Optional[PoolerOutput]:
-        hidden_states = hidden_states.float()
-        logits = self.score(hidden_states)
-        return self._pooler(logits, pooling_metadata)
-
-    def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
-        # TODO: The reward weights themselves have float32 accuracy data, we
-        # would like to load them in fp32 to get that extra precision.
-        super().load_weights(weights)
-        self.score = self.score.float()
+        return self._pooler(hidden_states, pooling_metadata)
diff --git a/vllm/model_executor/models/modernbert.py b/vllm/model_executor/models/modernbert.py
@@ -1,7 +1,7 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 from collections.abc import Iterable
-from typing import Optional
+from typing import Optional, Union
 
 import torch
 from torch import nn
@@ -13,7 +13,8 @@
 from vllm.distributed import get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.linear import (QKVParallelLinear,
                                                RowParallelLinear)
-from vllm.model_executor.layers.pooler import ClassifierPooler
+from vllm.model_executor.layers.pooler import (BasePooler, ClassifierPooler,
+                                               PoolingMethod, PoolingType)
 from vllm.model_executor.layers.rotary_embedding import RotaryEmbedding
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding)
@@ -252,10 +253,13 @@ def forward(
         return norm_outputs
 
 
-class ModernBertPooler(nn.Module):
+class ModernBertPooler(BasePooler):
 
     def __init__(self, config: ModernBertConfig):
         super().__init__()
+
+        pooling_type = PoolingType[config.classifier_pooling.upper()]
+        self.pooling = PoolingMethod.from_pooling_type(pooling_type)
         self.dense = nn.Linear(config.hidden_size, config.hidden_size,
                                config.classifier_bias)
         self.pooling_type = config.classifier_pooling
@@ -264,15 +268,12 @@ def __init__(self, config: ModernBertConfig):
                                  eps=config.norm_eps,
                                  bias=config.norm_bias)
 
-    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        pooled_output = hidden_states
-        if self.pooling_type == "mean":
-            pooled_output = pooled_output.mean(dim=0, keepdim=False)
-        elif self.pooling_type == "cls":
-            pooled_output = pooled_output[0, :]
-        else:
-            raise ValueError("Pooling type should be either `cls` or `mean`, "
-                             f"but got {self.pooling_type}")
+    def forward(
+        self,
+        hidden_states: Union[torch.Tensor, list[torch.Tensor]],
+        pooling_metadata: PoolingMetadata,
+    ) -> Union[torch.Tensor, list[torch.Tensor]]:
+        pooled_output = self.pooling(hidden_states, pooling_metadata)
         pooled_output = self.norm(self.act(self.dense(pooled_output)))
         return pooled_output
 
@@ -287,9 +288,11 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         self.model = ModernBertModel(vllm_config=vllm_config,
                                      prefix=maybe_prefix(prefix, "modernbert"))
         self.classifier = nn.Linear(config.hidden_size, config.num_labels)
-        self._pooler = ClassifierPooler(vllm_config.model_config,
-                                        self.classifier,
-                                        ModernBertPooler(config))
+        self._pooler = ClassifierPooler(
+            vllm_config.model_config,
+            pooling=ModernBertPooler(config),
+            classifier=self.classifier,
+        )
 
     def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
 
diff --git a/vllm/model_executor/models/roberta.py b/vllm/model_executor/models/roberta.py
@@ -9,7 +9,7 @@
 from transformers import RobertaConfig
 
 from vllm.config import VllmConfig
-from vllm.model_executor.layers.pooler import ClassifierPooler
+from vllm.model_executor.layers.pooler import ClassifierPooler, CLSPool
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     VocabParallelEmbedding)
 from vllm.model_executor.models.bert import BertEmbeddingModel, BertModel
@@ -106,8 +106,8 @@ def __init__(self, config: RobertaConfig):
         self.dense = nn.Linear(config.hidden_size, config.hidden_size)
         self.out_proj = nn.Linear(config.hidden_size, config.num_labels)
 
-    def forward(self, features, **kwargs):
-        x = features[0, :]  # take <s> token (equiv. to [CLS])
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        # CLSPool has already been applied in `pooling`
         x = self.dense(x)
         x = torch.tanh(x)
         x = self.out_proj(x)
@@ -188,8 +188,11 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
                                  add_pooling_layer=False)
         self.classifier = RobertaClassificationHead(config)
 
-        self._pooler = ClassifierPooler(vllm_config.model_config,
-                                        self.classifier)
+        self._pooler = ClassifierPooler(
+            vllm_config.model_config,
+            pooling=CLSPool(),
+            classifier=self.classifier,
+        )
 
     def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
         loader = AutoWeightsLoader(self)
diff --git a/vllm/transformers_utils/config.py b/vllm/transformers_utils/config.py
