[V1][TPU] TPU multimodal model support for ragged attention

vllm/v1/worker/tpu_model_runner.py

@@ -15,14 +15,18 @@
 from vllm.attention.layer import Attention
 from vllm.config import VllmConfig
 from vllm.forward_context import get_forward_context, set_forward_context
+from vllm.inputs import INPUT_REGISTRY
 from vllm.logger import init_logger
 from vllm.model_executor.model_loader import get_model
+from vllm.multimodal import MULTIMODAL_REGISTRY, MultiModalKwargs
+from vllm.multimodal.utils import group_mm_inputs_by_modality
 from vllm.sampling_params import SamplingType
 from vllm.utils import LayerBlockType, cdiv, is_pin_memory_available
 from vllm.v1.attention.backends.pallas import (NUM_KV_PAGES_PER_BLOCK,
                                                NUM_QUERIES_PER_BLOCK,
                                                PallasAttentionBackend,
                                                PallasMetadata)
+from vllm.v1.core.encoder_cache_manager import compute_encoder_budget
 from vllm.v1.kv_cache_interface import (FullAttentionSpec, KVCacheConfig,
                                         KVCacheSpec)
 from vllm.v1.outputs import LogprobsTensors, ModelRunnerOutput
@@ -72,8 +76,10 @@ def __init__(
         self.block_size = cache_config.block_size
         self.max_model_len = model_config.max_model_len
         self.max_num_blocks_per_req = cdiv(self.max_model_len, self.block_size)
-        self.max_num_tokens = scheduler_config.max_num_batched_tokens
-        self.max_num_reqs = scheduler_config.max_num_seqs
+        self.max_num_tokens = _get_padded_number(
+            scheduler_config.max_num_batched_tokens, NUM_QUERIES_PER_BLOCK)
+        self.max_num_reqs = _get_padded_number(scheduler_config.max_num_seqs,
+                                               NUM_QUERIES_PER_BLOCK)
 
         # Model-related.
         self.num_attn_layers = model_config.get_num_layers_by_block_type(
@@ -84,25 +90,38 @@ def __init__(
         self.head_size = model_config.get_head_size()
         self.hidden_size = model_config.get_hidden_size()
 
+        # Multi-modal data support
+        self.input_registry = INPUT_REGISTRY
+        self.mm_registry = MULTIMODAL_REGISTRY
+        self.uses_mrope = model_config.uses_mrope
+        # TODO: Support M-RoPE (e.g, Qwen2-VL)
+        assert not self.uses_mrope, "TPU does not support M-RoPE yet."
+
+        encoder_compute_budget, encoder_cache_size = compute_encoder_budget(
+            model_config=model_config,
+            scheduler_config=scheduler_config,
+        )
+        self.max_num_encoder_input_tokens = encoder_compute_budget
+        self.encoder_cache_size = encoder_cache_size
+
+        # Lazy initialization
+        # self.model: nn.Module  # Set after load_model
+        self.kv_caches: list[torch.Tensor] = []
+        # req_id -> (input_id -> encoder_output)
+        self.encoder_cache: dict[str, dict[int, torch.Tensor]] = {}
+
+        # Request states.
+        self.requests: dict[str, CachedRequestState] = {}
         # Persistent batch.
         self.input_batch = InputBatch(
             max_num_reqs=self.max_num_reqs,
             max_model_len=self.max_model_len,
             max_num_blocks_per_req=self.max_num_blocks_per_req,
             device=self.device,
             pin_memory=self.pin_memory,
-            vocab_size=self.model_config.get_vocab_size(),
+            vocab_size=model_config.get_vocab_size(),
         )
 
-        # Request states.
-        self.requests: dict[str, CachedRequestState] = {}
-
-        # req_id -> (input_id -> encoder_output)
-        self.encoder_cache: dict[str, dict[int, torch.Tensor]] = {}
-
-        # KV caches for forward pass
-        self.kv_caches: list[tuple[torch.Tensor, torch.Tensor]] = []
-
         # Cached torch/numpy tensor
         # The pytorch tensor and numpy array share the same buffer.
         # Sometimes the numpy op is faster so we create both.
@@ -164,6 +183,7 @@ def _update_states(self, scheduler_output: "SchedulerOutput") -> bool:
         # Remove finished requests from the cached states.
         for req_id in scheduler_output.finished_req_ids:
             self.requests.pop(req_id, None)
+            self.encoder_cache.pop(req_id, None)
 
         # Remove the finished requests from the persistent batch.
         # NOTE(woosuk): There could be an edge case where finished_req_ids and
@@ -177,6 +197,14 @@ def _update_states(self, scheduler_output: "SchedulerOutput") -> bool:
             if req_index is not None:
                 removed_req_indices.append(req_index)
 
+        # Free the cached encoder outputs.
+        for req_id, input_id in scheduler_output.free_encoder_input_ids:
+            encoder_outputs = self.encoder_cache.get(req_id)
+            if encoder_outputs is not None:
+                encoder_outputs.pop(input_id, None)
+                if not encoder_outputs:
+                    self.encoder_cache.pop(req_id, None)
+
         # Remove the unscheduled requests from the persistent batch.
         # NOTE(woosuk): The unscheduled requests are either preempted requests
         # or running requests that are not scheduled in this step. We remove
@@ -426,6 +454,92 @@ def _prepare_inputs(self, scheduler_output: "SchedulerOutput"):
         logits_indices = query_start_loc[1:] - 1
         return attn_metadata, logits_indices
 
+    def _execute_encoder(self, scheduler_output: "SchedulerOutput"):
+        scheduled_encoder_inputs = scheduler_output.scheduled_encoder_inputs
+        if not scheduled_encoder_inputs:
+            return
+
+        # Batch the multi-modal inputs.
+        mm_inputs: list[MultiModalKwargs] = []
+        req_input_ids: list[tuple[str, int]] = []
+        for req_id, encoder_input_ids in scheduled_encoder_inputs.items():
+            req_state = self.requests[req_id]
+            for input_id in encoder_input_ids:
+                mm_inputs.append(req_state.mm_inputs[input_id])
+                req_input_ids.append((req_id, input_id))
+
+        # Batch mm inputs as much as we can: if a request in the batch has
+        # multiple modalities or a different modality than the previous one,
+        # we process it separately to preserve item order.
+        # FIXME(ywang96): This is a hacky way to deal with multiple modalities
+        # in the same batch while still being able to benefit from batching
+        # multimodal inputs. The proper solution should be reordering the
+        # encoder outputs.
+        grouped_mm_inputs_list = group_mm_inputs_by_modality(mm_inputs)
+
+        encoder_outputs = []
+        for grouped_mm_inputs in grouped_mm_inputs_list:
+            batched_mm_inputs = MultiModalKwargs.batch(grouped_mm_inputs)
+            batched_mm_inputs = MultiModalKwargs.as_kwargs(batched_mm_inputs,
+                                                           device=self.device)
+
+            # Run the encoder.
+            # `curr_group_outputs` is either of the following:
+            # 1. A tensor of shape (num_items, feature_size, hidden_size)
+            # in case feature_size is fixed across all multimodal items.
+            # 2. A list or tuple (length: num_items) of tensors, each of shape
+            # (feature_size, hidden_size) in case the feature size is dynamic
+            # depending on the input multimodal items.
+            curr_group_outputs = self.model.get_multimodal_embeddings(
+                **batched_mm_inputs)
+
+            for output in curr_group_outputs:
+                encoder_outputs.append(output)
+
+        # Cache the encoder outputs.
+        for (req_id, input_id), output in zip(req_input_ids, encoder_outputs):
+            if req_id not in self.encoder_cache:
+                self.encoder_cache[req_id] = {}
+            self.encoder_cache[req_id][input_id] = output
+
+    def _gather_encoder_outputs(
+        self,
+        scheduler_output: "SchedulerOutput",
+    ) -> list[torch.Tensor]:
+        encoder_outputs: list[torch.Tensor] = []
+        for req_id in self.input_batch.req_ids:
+            num_scheduled_tokens = scheduler_output.num_scheduled_tokens[
+                req_id]
+            req_state = self.requests[req_id]
+            num_computed_tokens = req_state.num_computed_tokens
+            mm_positions = req_state.mm_positions
+            for i, pos_info in enumerate(mm_positions):
+                start_pos = pos_info["offset"]
+                num_encoder_tokens = pos_info["length"]
+
+                # The encoder output is needed if the two ranges overlap:
+                # [num_computed_tokens,
+                #  num_computed_tokens + num_scheduled_tokens) and
+                # [start_pos, start_pos + num_encoder_tokens)
+                if start_pos >= num_computed_tokens + num_scheduled_tokens:
+                    # The encoder output is not needed in this step.
+                    break
+                if start_pos + num_encoder_tokens <= num_computed_tokens:
+                    # The encoder output is already processed and stored
+                    # in the decoder's KV cache.
+                    continue
+
+                start_idx = max(num_computed_tokens - start_pos, 0)
+                end_idx = min(
+                    num_computed_tokens - start_pos + num_scheduled_tokens,
+                    num_encoder_tokens)
+                assert start_idx < end_idx
+                assert req_id in self.encoder_cache
+                assert i in self.encoder_cache[req_id]
+                encoder_output = self.encoder_cache[req_id][i]
+                encoder_outputs.append(encoder_output[start_idx:end_idx])
+        return encoder_outputs
+
     @torch.no_grad()
     def execute_model(
         self,
@@ -434,16 +548,42 @@ def execute_model(
         # Update cached state
         self._update_states(scheduler_output)
 
+        if self.is_multimodal_model:
+            # Run the multimodal encoder if any.
+            self._execute_encoder(scheduler_output)
+            encoder_outputs = self._gather_encoder_outputs(scheduler_output)
+        else:
+            encoder_outputs = []
+
         # Prepare inputs
         attn_metadata, logits_indices = self._prepare_inputs(scheduler_output)
         total_num_scheduled_tokens = scheduler_output.total_num_scheduled_tokens
 
+        if self.is_multimodal_model:
+            # NOTE(woosuk): To unify token ids and soft tokens (vision
+            # embeddings), we always use embeddings (rather than token ids)
+            # as input to the multimodal model, even when the input is text.
+            if encoder_outputs:
+                inputs_embeds = self.model.get_input_embeddings(
+                    self.input_ids, encoder_outputs)
+            else:
+                inputs_embeds = self.model.get_input_embeddings(self.input_ids)
+            input_ids = None
+        else:
+            # For text-only models, we use token ids as input.
+            # While it is possible to use embeddings as input just like the
+            # multimodal models, it is not desirable for performance since
+            # then the embedding layer is not included in the CUDA graph.
+            input_ids = self.input_ids
+            inputs_embeds = None
+
         # Run the decoder
         with set_forward_context(attn_metadata, self.vllm_config):
             hidden_states = self.model(
-                token_ids=self.input_ids,
-                position_ids=self.position_ids,
+                input_ids=input_ids,
+                positions=self.position_ids,
                 kv_caches=self.kv_caches,
+                inputs_embeds=inputs_embeds,
             )
         hidden_states = hidden_states[:total_num_scheduled_tokens]
         num_reqs = self.input_batch.num_reqs
@@ -538,14 +678,21 @@ def load_model(self) -> None:
                                    fullgraph=True,
                                    dynamic=False)
 
-    def dummy_run(
+    def _dummy_run(
         self,
         kv_caches,
         num_tokens: int,
     ) -> None:
-        input_ids = torch.zeros(num_tokens,
-                                dtype=torch.int32,
-                                device=self.device)
+        if self.is_multimodal_model:
+            input_ids = None
+            inputs_embeds = torch.zeros((num_tokens, self.hidden_size),
+                                        dtype=self.dtype,
+                                        device=self.device)
+        else:
+            input_ids = torch.zeros((num_tokens),
+                                    dtype=torch.int32,
+                                    device=self.device)
+            inputs_embeds = None
         position_ids = torch.zeros(num_tokens,
                                    dtype=torch.int32,
                                    device=self.device)
@@ -571,7 +718,10 @@ def dummy_run(
             num_seqs=num_tokens,
         )
 
-        torch._dynamo.mark_dynamic(input_ids, 0)
+        if self.is_multimodal_model:
+            torch._dynamo.mark_dynamic(inputs_embeds, 0)
+        else:
+            torch._dynamo.mark_dynamic(input_ids, 0)
         torch._dynamo.mark_dynamic(position_ids, 0)
         torch._dynamo.mark_dynamic(attn_metadata.slot_mapping, 0)
         torch._dynamo.mark_dynamic(attn_metadata.block_tables, 0)
@@ -580,7 +730,12 @@ def dummy_run(
 
         with set_forward_context(attn_metadata, self.vllm_config, 0):
             assert self.model is not None
-            self.model(input_ids, position_ids, kv_caches)
+            self.model(
+                input_ids=input_ids,
+                positions=position_ids,
+                kv_caches=kv_caches,
+                inputs_embeds=inputs_embeds,
+            )
 
     def capture_model(self) -> None:
         """Compile the model."""
@@ -590,11 +745,11 @@ def capture_model(self) -> None:
         start = time.perf_counter()
         num_tokens = 16
         while True:
-            self.dummy_run(self.kv_caches, num_tokens)
+            self._dummy_run(self.kv_caches, num_tokens)
             logger.info("  -- num_tokens: %d", num_tokens)
             xm.mark_step()
             xm.wait_device_ops()
-            if num_tokens >= self.scheduler_config.max_num_batched_tokens:
+            if num_tokens >= self.max_num_tokens:
                 break
             num_tokens *= 2
         end = time.perf_counter()
@@ -647,17 +802,20 @@ def __init__(self, model: nn.Module):
 
     def forward(
         self,
-        token_ids: torch.Tensor,
-        position_ids: torch.Tensor,
+        input_ids: torch.Tensor,
+        positions: torch.Tensor,
         kv_caches: list[tuple[torch.Tensor, torch.Tensor]],
+        inputs_embeds: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
         """Executes the forward pass of the model and samples the next token.
 
         Args:
-            token_ids: The input token IDs of shape [num_tokens].
-            position_ids: The input position IDs of shape [num_tokens].
+            input_ids: The input token IDs of shape [num_tokens].
+            positions: The input position IDs of shape [num_tokens].
             kv_caches: The key and value caches. They can be None during the
                 memory profiling at initialization.
+            inputs_embeds: The input embeddings of shape [num_tokens,
+                hidden_size]. It is used for multimodal models.
         """
         # Skip this in memory profiling at initialization.
         if kv_caches[0][0].numel() > 0:
@@ -684,9 +842,9 @@ def forward(
 
         assert self.model is not None
         hidden_states = self.model(
-            token_ids,
-            position_ids,
-            kv_caches,
+            input_ids=input_ids,
+            positions=positions,
+            inputs_embeds=inputs_embeds,
         )
 
         return hidden_states
@@ -699,6 +857,12 @@ def compute_logits(
         logits = self.model.compute_logits(hidden_states, sampling_metadata)
         return logits
 
+    def get_multimodal_embeddings(self, *args, **kwargs):
+        return self.model.get_multimodal_embeddings(*args, **kwargs)
+
+    def get_input_embeddings(self, *args, **kwargs):
+        return self.model.get_input_embeddings(*args, **kwargs)
+
 
 def _get_padded_number(n: int, multiple: int) -> int:
     return ((n + multiple - 1) // multiple) * multiple

vllm/v1/worker/tpu_worker.py

@@ -124,7 +124,7 @@ def determine_available_memory(self) -> int:
             self.vllm_config.compilation_config.static_forward_context,
             runner_kv_caches)
 
-        self.model_runner.dummy_run(
+        self.model_runner._dummy_run(
             runner_kv_caches,
             num_tokens=self.scheduler_config.max_num_batched_tokens,
         )