[Cleanup] Remove unused attn_metadata parameter from Proposer classes (#4862)

The `attn_metadata` is not used by any draft proposer, so we can remove
it.


- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>

tests/ut/spec_decode/test_mtp_proposer.py

@@ -238,7 +238,6 @@ class TestMtpProposer:
         proposer.speculative_config = MagicMock(
             disable_padded_drafter_batch=False)
         proposer.pcp_size = mock_deps.runner.pcp_size
-        proposer._get_attn_metadata = MagicMock(return_value=MagicMock())
         proposer.prepare_next_token_ids_padded = MagicMock(
             return_value=(torch.tensor([101, 200, 302]), 3))
         proposer.prepare_inputs_padded = MagicMock(
@@ -261,7 +260,6 @@ class TestMtpProposer:
 
         proposer.prepare_next_token_ids_padded.assert_called_once()
         proposer.prepare_inputs_padded.assert_called_once()
-        proposer._get_attn_metadata.assert_called_once()
         proposer._propose.assert_called_once()
         assert torch.equal(draft_token_ids, proposer._propose.return_value)
 

vllm_ascend/spec_decode/eagle_proposer.py

@@ -144,7 +144,6 @@ class EagleProposer(Proposer):
                            positions: torch.Tensor = None,
                            num_scheduled_tokens: int = 0,
                            hidden_states: torch.Tensor = None,
-                           attn_metadata=None,
                            aux_hidden_states: torch.Tensor = None):
 
         attn_metadata = self._get_eagle_atten_dict(scheduler_output)

vllm_ascend/spec_decode/interface.py

@@ -48,7 +48,6 @@ class Proposer:
                            positions: torch.Tensor = None,
                            num_scheduled_tokens: int = 0,
                            hidden_states: torch.Tensor = None,
-                           attn_metadata=None,
                            aux_hidden_states: torch.Tensor = None):
         """Called by execute_model in model_runner"""
         raise NotImplementedError

vllm_ascend/spec_decode/mtp_proposer.py

@@ -51,10 +51,6 @@ _MTP_MODELS = {
     ("vllm.model_executor.models.qwen3_next_mtp", "Qwen3NextMTP")
 }
 
-_DEFAULT_FIRST_LAYER = 'model.layers.0.self_attn.attn'
-
-_FIRST_LAYERS = {"Qwen3NextForCausalLM": 'model.layers.3.self_attn.attn'}
-
 
 def _load_model(architecture):
     if architecture not in _MTP_MODELS:
@@ -345,10 +341,8 @@ class MtpProposer(Proposer):
                            positions: torch.Tensor = None,
                            num_scheduled_tokens: int = 0,
                            hidden_states: torch.Tensor = None,
-                           attn_metadata=None,
                            aux_hidden_states: torch.Tensor = None):
         common_attn_metadata = self.runner.spec_decode_common_attn_metadata
-        attn_metadata = self._get_attn_metadata(attn_metadata)
 
         if self.speculative_config.disable_padded_drafter_batch:
             # When padded-batch is disabled, the sampled_token_ids should be
@@ -487,14 +481,6 @@ class MtpProposer(Proposer):
         model = _load_model(architecture)
         self.model = model(vllm_config=self.vllm_config).to(target_device)
 
-    def _get_attn_metadata(self, attn_metadata):
-        if attn_metadata is not None and isinstance(attn_metadata, dict):
-            architecture = self.vllm_config.model_config.architecture
-            layer_name = _FIRST_LAYERS.get(architecture, _DEFAULT_FIRST_LAYER)
-            attn_metadata = attn_metadata[layer_name]
-
-        return attn_metadata
-
     def _prepare_inputs(
         self,
         common_attn_metadata: CommonAttentionMetadata,

vllm_ascend/spec_decode/ngram_proposer.py

@@ -38,7 +38,6 @@ class NgramProposer(VllmNgramProposer, Proposer):
                            positions=None,
                            num_scheduled_tokens=None,
                            hidden_states=None,
-                           attn_metadata=None,
                            aux_hidden_states=None) -> list[list[int]]:
         valid_ngram_requests = []
         for i, sampled_ids in enumerate(valid_sampled_token_ids):

vllm_ascend/spec_decode/suffix_proposer.py

@@ -38,7 +38,6 @@ class SuffixDecodingProposer(VllmSuffixDecodingProposer, Proposer):
                            positions=None,
                            num_scheduled_tokens=None,
                            hidden_states=None,
-                           attn_metadata=None,
                            aux_hidden_states=None) -> list[list[int]]:
         draft_token_ids = self.propose(self.runner.input_batch,
                                        valid_sampled_token_ids)

vllm_ascend/worker/model_runner_v1.py

@@ -1383,7 +1383,7 @@ class NPUModelRunner(GPUModelRunner):
             draft_token_ids = self.drafter.generate_token_ids(
                 valid_sampled_token_ids, sampling_metadata, scheduler_output,
                 spec_decode_metadata, positions, num_scheduled_tokens,
-                hidden_states, attn_metadata, aux_hidden_states)
+                hidden_states, aux_hidden_states)
         return draft_token_ids
 
     def _select_moe_comm_method(self,