sglang 0.4.4__py3-none-any.whl → 0.4.4.post2__py3-none-any.whl

sglang/srt/models/qwen2_5_vl.py

@@ -26,7 +26,6 @@ import logging
 from functools import lru_cache, partial
 from typing import Iterable, List, Optional, Tuple, Type
 
-import numpy as np
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
@@ -34,8 +33,15 @@ from einops import rearrange
 from transformers import AutoModel, Qwen2VLConfig
 from transformers.activations import ACT2FN
 from transformers.models.qwen2.modeling_qwen2 import Qwen2RMSNorm
+from transformers.models.qwen2_5_vl import Qwen2_5_VLProcessor
+from transformers.models.qwen2_5_vl.configuration_qwen2_5_vl import (
+    Qwen2_5_VLConfig,
+    Qwen2_5_VLVisionConfig,
+)
+from transformers.models.qwen2_5_vl.modeling_qwen2_5_vl import (
+    Qwen2_5_VLForConditionalGeneration,
+)
 
-from sglang.srt.configs import Qwen2_5_VLConfig, Qwen2_5_VLVisionConfig
 from sglang.srt.distributed import (
     get_tensor_model_parallel_rank,
     get_tensor_model_parallel_world_size,
@@ -47,14 +53,15 @@ from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.pooler import Pooler, PoolingType
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.vocab_parallel_embedding import ParallelLMHead
-from sglang.srt.managers.multi_modality_padding import (
+from sglang.srt.managers.mm_utils import (
     MultiModalityDataPaddingPatternTokenPairs,
+    general_mm_embed_routine,
 )
-from sglang.srt.managers.schedule_batch import ImageInputs
+from sglang.srt.managers.schedule_batch import MultimodalInputs
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
 from sglang.srt.model_loader.weight_utils import default_weight_loader
 from sglang.srt.models.qwen2 import Qwen2Model
-from sglang.srt.models.qwen2_vl import Qwen2VLImageInputs, Qwen2VLVideoInputs
+from sglang.srt.models.qwen2_vl import Qwen2VLVideoInputs
 from sglang.srt.utils import add_prefix
 
 logger = logging.getLogger(__name__)
@@ -125,12 +132,15 @@ class Qwen2_5_VisionBlock(nn.Module):
         if attn_implementation == "sdpa":
             use_context_forward = False
             softmax_in_single_precision = False
+            flatten_batch = True
         elif attn_implementation == "flash_attention_2":
             softmax_in_single_precision = False
             use_context_forward = True
+            flatten_batch = True
         elif attn_implementation == "eager":
             softmax_in_single_precision = True
             use_context_forward = False
+            flatten_batch = True
 
         self.attn = VisionAttention(
             embed_dim=dim,
@@ -139,7 +149,7 @@ class Qwen2_5_VisionBlock(nn.Module):
             use_qkv_parallel=False,
             use_context_forward=use_context_forward,
             softmax_in_single_precision=softmax_in_single_precision,
-            flatten_batch=True,
+            flatten_batch=flatten_batch,
             quant_config=quant_config,
             prefix=add_prefix("attn", prefix),
         )
@@ -192,9 +202,10 @@ class Qwen2_5_VisionPatchEmbed(nn.Module):
         )
 
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        target_dtype = self.proj.weight.dtype
         L, C = x.shape
         x = x.view(L, -1, self.temporal_patch_size, self.patch_size, self.patch_size)
-        x = self.proj(x).view(L, self.embed_dim)
+        x = self.proj(x.to(dtype=target_dtype)).view(L, self.embed_dim)
         return x
 
 
@@ -246,35 +257,15 @@ class Qwen2_5_VisionRotaryEmbedding(nn.Module):
 
     def __init__(self, dim: int, theta: float = 10000.0) -> None:
         super().__init__()
-        self.dim = dim
-        self.theta = theta
         inv_freq = 1.0 / (theta ** (torch.arange(0, dim, 2, dtype=torch.float) / dim))
         self.register_buffer("inv_freq", inv_freq, persistent=False)
-        self._seq_len_cached = 0
-        self._freqs_cached = None
-
-    def update_freqs_cache(self, seqlen: int) -> None:
-        if seqlen > self._seq_len_cached:
-            seqlen *= 2
-            self._seq_len_cached = seqlen
-            self.inv_freq = 1.0 / (
-                self.theta
-                ** (
-                    torch.arange(
-                        0, self.dim, 2, dtype=torch.float, device=self.inv_freq.device
-                    )
-                    / self.dim
-                )
-            )
-            seq = torch.arange(
-                seqlen, device=self.inv_freq.device, dtype=self.inv_freq.dtype
-            )
-            freqs = torch.outer(seq, self.inv_freq)
-            self._freqs_cached = freqs
 
     def forward(self, seqlen: int) -> torch.Tensor:
-        self.update_freqs_cache(seqlen)
-        return self._freqs_cached[:seqlen]
+        seq = torch.arange(
+            seqlen, device=self.inv_freq.device, dtype=self.inv_freq.dtype
+        )
+        freqs = torch.outer(seq, self.inv_freq)
+        return freqs
 
 
 class Qwen2_5_VisionTransformer(nn.Module):
@@ -293,7 +284,7 @@ class Qwen2_5_VisionTransformer(nn.Module):
         spatial_merge_size: int = vision_config.spatial_merge_size
         self.spatial_merge_size = spatial_merge_size
         self.spatial_merge_unit: int = spatial_merge_size * spatial_merge_size
-        in_chans: int = vision_config.in_chans
+        in_chans: int = vision_config.in_channels
         hidden_size: int = vision_config.hidden_size
         depth: int = vision_config.depth
         num_heads: int = vision_config.num_heads
@@ -335,13 +326,12 @@ class Qwen2_5_VisionTransformer(nn.Module):
         )
 
     def get_window_index(self, grid_thw):
-        window_index: list = []
         cu_window_seqlens: list = [0]
         window_index_id = 0
         vit_merger_window_size = (
             self.window_size // self.spatial_merge_size // self.patch_size
         )
-
+        window_index: list = []
         for grid_t, grid_h, grid_w in grid_thw:
             llm_grid_h, llm_grid_w = (
                 grid_h // self.spatial_merge_size,
@@ -378,7 +368,6 @@ class Qwen2_5_VisionTransformer(nn.Module):
             cu_window_seqlens.extend(cu_seqlens_tmp.tolist())
             window_index_id += (grid_t * llm_grid_h * llm_grid_w).item()
         window_index = torch.cat(window_index, dim=0)
-
         return window_index, cu_window_seqlens
 
     @property
@@ -391,29 +380,29 @@ class Qwen2_5_VisionTransformer(nn.Module):
 
     def rot_pos_emb(self, grid_thw: torch.Tensor) -> torch.Tensor:
         pos_ids = []
-        for t, h, w in grid_thw:
+        for i in range(grid_thw.size(0)):
+            t, h, w = grid_thw[i].tolist()
             hpos_ids = torch.arange(h).unsqueeze(1).expand(-1, w)
-            wpos_ids = torch.arange(w).unsqueeze(0).expand(h, -1)
-            hpos_ids = (
-                hpos_ids.reshape(
-                    h // self.spatial_merge_size,
-                    self.spatial_merge_size,
-                    w // self.spatial_merge_size,
-                    self.spatial_merge_size,
-                )
-                .permute(0, 2, 1, 3)
-                .flatten()
+
+            hpos_ids = hpos_ids.reshape(
+                h // self.spatial_merge_size,
+                self.spatial_merge_size,
+                w // self.spatial_merge_size,
+                self.spatial_merge_size,
             )
-            wpos_ids = (
-                wpos_ids.reshape(
-                    h // self.spatial_merge_size,
-                    self.spatial_merge_size,
-                    w // self.spatial_merge_size,
-                    self.spatial_merge_size,
-                )
-                .permute(0, 2, 1, 3)
-                .flatten()
+            hpos_ids = hpos_ids.permute(0, 2, 1, 3)
+            hpos_ids = hpos_ids.flatten()
+
+            wpos_ids = torch.arange(w).unsqueeze(0).expand(h, -1)
+            wpos_ids = wpos_ids.reshape(
+                h // self.spatial_merge_size,
+                self.spatial_merge_size,
+                w // self.spatial_merge_size,
+                self.spatial_merge_size,
             )
+            wpos_ids = wpos_ids.permute(0, 2, 1, 3)
+            wpos_ids = wpos_ids.flatten()
+
             pos_ids.append(torch.stack([hpos_ids, wpos_ids], dim=-1).repeat(t, 1))
         pos_ids = torch.cat(pos_ids, dim=0)
         max_grid_size = grid_thw[:, 1:].max()
@@ -437,7 +426,7 @@ class Qwen2_5_VisionTransformer(nn.Module):
         cu_window_seqlens = torch.tensor(
             cu_window_seqlens,
             device=x.device,
-            dtype=grid_thw.dtype if torch.jit.is_tracing() else torch.int32,
+            dtype=torch.int32,
         )
         cu_window_seqlens = torch.unique_consecutive(cu_window_seqlens)
 
@@ -455,9 +444,12 @@ class Qwen2_5_VisionTransformer(nn.Module):
         position_embeddings = (emb.cos(), emb.sin())
 
         # compute cu_seqlens
-        cu_seqlens = torch.repeat_interleave(
-            grid_thw[:, 1] * grid_thw[:, 2], grid_thw[:, 0]
-        ).cumsum(dim=0, dtype=torch.int32)
+        cu_seqlens = torch.cat(
+            [
+                torch.tensor([0], device=grid_thw.device),
+                (grid_thw[:, 0] * grid_thw[:, 1] * grid_thw[:, 2]).cumsum(dim=0),
+            ]
+        )
         cu_seqlens = F.pad(cu_seqlens, (1, 0), "constant", 0)
 
         # transformers
@@ -521,19 +513,7 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module):
         self.logits_processor = LogitsProcessor(config)
         self.pooler = Pooler(pooling_type=PoolingType.LAST, normalize=True)
 
-    def calculate_num_image_tokens(self, image_grid_thw: Tuple[int, int, int]):
-        processor = cached_get_processor(self.config._name_or_path)
-        grid_t, grid_h, grid_w = image_grid_thw
-        num_image_tokens = (
-            grid_t
-            * grid_h
-            * grid_w
-            // processor.image_processor.merge_size
-            // processor.image_processor.merge_size
-        )
-        return num_image_tokens
-
-    def pad_input_ids(self, input_ids: List[int], image_inputs: ImageInputs):
+    def pad_input_ids(self, input_ids: List[int], image_inputs: MultimodalInputs):
         # Get all special token IDs
         im_start_id: int = image_inputs.im_start_id
         im_end_id: int = image_inputs.im_end_id
@@ -543,9 +523,9 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module):
 
         return pattern.pad_input_tokens(input_ids, image_inputs)
 
-    def _process_image_input(self, image_input: Qwen2VLImageInputs) -> torch.Tensor:
-        pixel_values = image_input["pixel_values"].type(self.visual.dtype)
-        image_embeds = self.visual(pixel_values, grid_thw=image_input["image_grid_thw"])
+    def get_image_feature(self, image_input: MultimodalInputs) -> torch.Tensor:
+        pixel_values = image_input.pixel_values.type(self.visual.dtype)
+        image_embeds = self.visual(pixel_values, grid_thw=image_input.image_grid_thws)
         return image_embeds
 
     def _process_video_input(self, video_input: Qwen2VLVideoInputs) -> torch.Tensor:
@@ -555,6 +535,9 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module):
         )
         return video_embeds
 
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
     def forward(
         self,
         input_ids: torch.Tensor,
@@ -577,85 +560,25 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module):
         if getattr(self.config, "rope_scaling", {}).get("type", None) == "mrope":
             positions = forward_batch.mrope_positions
 
-        image_inputs = None
-        if forward_batch.image_inputs is not None:
-            image_inputs = [
-                img for img in forward_batch.image_inputs if img is not None
-            ]
-
-        if (
+        if not (
             forward_batch.forward_mode.is_decode()
-            or image_inputs is None
-            or len(image_inputs) == 0
+            or not forward_batch.contains_image_inputs()
         ):
-            inputs_embeds = self.model.embed_tokens(input_ids)
-        else:
             if getattr(self.config, "rope_scaling", {}).get("type", None) == "mrope":
                 assert positions.ndim == 2 and positions.size(0) == 3, (
                     "multimodal section rotary embedding requires "
                     f"(3, seq_len) positions, but got {positions.size()}"
                 )
 
-            # Clamp input ids. This is because the input_ids for the image tokens are
-            # filled with the hash values of the image for the prefix matching in the radix attention.
-            # There values are useless because their embeddings will be replaced by vision embeddings anyway.
-            input_ids.clamp_(min=0, max=self.config.vocab_size - 1)
-            # [B, s, hidden_size]
-            inputs_embeds = self.model.embed_tokens(input_ids)
-            extend_start_loc_cpu = forward_batch.extend_start_loc.cpu().numpy()
-            prefix_lens_cpu = forward_batch.extend_prefix_lens_cpu
-            for i, image in enumerate(forward_batch.image_inputs):
-                if image is None or image.pixel_values is None:
-                    continue
-                start_idx = extend_start_loc_cpu[i]
-                prefix_len = prefix_lens_cpu[i]
-
-                pixel_values = image.pixel_values.clone().detach().requires_grad_(False)
-                image_grid_thws = torch.tensor(
-                    np.array(image.image_grid_thws), device="cuda"
-                )
-                image_offsets = image.image_offsets
-                image_input = Qwen2VLImageInputs(
-                    pixel_values=pixel_values, image_grid_thw=image_grid_thws
-                )
-                image_embeds = self._process_image_input(image_input)
-
-                image_embeds_offset = 0
-                for idx, image_offset in enumerate(image_offsets):
-                    if image_offset < prefix_len:
-                        continue
-                    num_image_tokens = self.calculate_num_image_tokens(
-                        image_grid_thws[idx]
-                    )
-
-                    left_idx = start_idx + (image_offset - prefix_len)
-                    right_idx = left_idx + num_image_tokens
-
-                    tp_size = get_tensor_model_parallel_world_size()
-
-                    hidden_size = image_embeds.shape[-1]
-
-                    if hidden_size % tp_size != 0:
-                        padding_size = tp_size - (hidden_size % tp_size)
-                        image_embeds = F.pad(image_embeds, (0, padding_size))
-                        inputs_embeds = F.pad(inputs_embeds, (0, padding_size))
-
-                    hidden_chunk_size = image_embeds.shape[-1] // tp_size
-                    rank = get_tensor_model_parallel_rank()
-                    start_dim = rank * hidden_chunk_size
-                    end_dim = (rank + 1) * hidden_chunk_size
-                    inputs_embeds[left_idx:right_idx, ..., start_dim:end_dim] = (
-                        image_embeds[
-                            image_embeds_offset : image_embeds_offset
-                            + num_image_tokens,
-                            ...,
-                            start_dim:end_dim,
-                        ]
-                    )
-                    image_embeds_offset += num_image_tokens
+        inputs_embeds = general_mm_embed_routine(
+            input_ids=input_ids,
+            forward_batch=forward_batch,
+            embed_tokens=self.get_input_embeddings(),
+            mm_data_embedding_func=self.get_image_feature,
+        )
 
         hidden_states = self.model(
-            input_ids=input_ids,
+            input_ids=None,
             positions=positions,
             forward_batch=forward_batch,
             input_embeds=inputs_embeds,
@@ -732,4 +655,3 @@ class Qwen2_5_VLForConditionalGeneration(nn.Module):
 
 
 EntryClass = [Qwen2_5_VLForConditionalGeneration]
-AutoModel.register(Qwen2_5_VLConfig, Qwen2_5_VLForConditionalGeneration)

sglang/srt/models/qwen2_classification.py

@@ -0,0 +1,75 @@
+# Copyright 2023-2024 SGLang Team
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+
+from typing import Iterable, Optional, Tuple
+
+import torch
+from torch import nn
+from transformers import Qwen2Config
+
+from sglang.srt.layers.pooler import EmbeddingPoolerOutput, Pooler, PoolingType
+from sglang.srt.layers.quantization.base_config import QuantizationConfig
+from sglang.srt.model_executor.forward_batch_info import ForwardBatch
+from sglang.srt.models.qwen2 import Qwen2ForCausalLM, Qwen2Model
+from sglang.srt.utils import add_prefix
+
+
+class Qwen2ForSequenceClassification(nn.Module):
+    def __init__(
+        self,
+        config: Qwen2Config,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+    ) -> None:
+        super().__init__()
+        self.config = config
+        self.quant_config = quant_config
+        self.model = Qwen2Model(
+            config, quant_config=quant_config, prefix=add_prefix("model", prefix)
+        )
+        self.score = nn.Linear(config.hidden_size, config.num_labels)
+        self.pooler = Pooler(pooling_type=PoolingType.LAST, normalize=False)
+
+        self.eos_token_id = config.eos_token_id
+
+    @torch.no_grad()
+    def forward(
+        self,
+        input_ids: torch.Tensor,
+        positions: torch.Tensor,
+        forward_batch: ForwardBatch,
+        input_embeds: torch.Tensor = None,
+        get_embedding: bool = True,
+    ) -> EmbeddingPoolerOutput:
+        assert (
+            get_embedding
+        ), "Qwen2ForSequenceClassification is only used for embedding"
+
+        hidden_states = self.model(input_ids, positions, forward_batch, input_embeds)
+        logits = self.score(hidden_states)
+        pooled_logits = self.pooler(logits, forward_batch).embeddings
+
+        return EmbeddingPoolerOutput(pooled_logits)
+
+    def load_weights(self, weights: Iterable[Tuple[str, torch.Tensor]]):
+        # Filter out lm_head weights of Qwen2ForCausalLM
+        filtered_weights = [
+            (name, w) for name, w in weights if not name.startswith("lm_head")
+        ]
+        return Qwen2ForCausalLM.load_weights(self, filtered_weights)
+
+
+EntryClass = [
+    Qwen2ForSequenceClassification,
+]

sglang/srt/models/qwen2_moe.py

@@ -44,10 +44,13 @@ from sglang.srt.layers.vocab_parallel_embedding import (
     ParallelLMHead,
     VocabParallelEmbedding,
 )
+from sglang.srt.managers.expert_distribution import ExpertDistributionRecorder
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
 from sglang.srt.model_loader.weight_utils import default_weight_loader
 from sglang.srt.utils import add_prefix
 
+expert_distribution_recorder = ExpertDistributionRecorder()
+
 
 class Qwen2MoeMLP(nn.Module):
     def __init__(
@@ -170,6 +173,7 @@ class Qwen2MoeAttention(nn.Module):
         rope_theta: float = 10000,
         rope_scaling: Optional[Dict[str, Any]] = None,
         max_position_embeddings: int = 8192,
+        qkv_bias: int = True,
         quant_config: Optional[QuantizationConfig] = None,
         prefix: str = "",
     ) -> None:
@@ -201,7 +205,7 @@ class Qwen2MoeAttention(nn.Module):
             self.head_dim,
             self.total_num_heads,
             self.total_num_kv_heads,
-            bias=True,
+            bias=qkv_bias,
             quant_config=quant_config,
             prefix=add_prefix("qkv_proj", prefix),
         )
@@ -257,6 +261,8 @@ class Qwen2MoeDecoderLayer(nn.Module):
         rope_theta = getattr(config, "rope_theta", 10000)
         rope_scaling = getattr(config, "rope_scaling", None)
         max_position_embeddings = getattr(config, "max_position_embeddings", 8192)
+        # note: replace config.num_hidden_layers < 80 with True once its available in transformers 4.50.0
+        qkv_bias = getattr(config, "qkv_bias", config.num_hidden_layers < 80)
         self.self_attn = Qwen2MoeAttention(
             hidden_size=self.hidden_size,
             num_heads=config.num_attention_heads,
@@ -266,6 +272,7 @@ class Qwen2MoeDecoderLayer(nn.Module):
             rope_scaling=rope_scaling,
             max_position_embeddings=max_position_embeddings,
             quant_config=quant_config,
+            qkv_bias=qkv_bias,
             prefix=add_prefix("self_attn", prefix),
         )
 
@@ -362,6 +369,7 @@ class Qwen2MoeModel(nn.Module):
             hidden_states = input_embeds
         residual = None
         for i in range(len(self.layers)):
+            expert_distribution_recorder.set_current_layer(i)
             layer = self.layers[i]
             hidden_states, residual = layer(
                 positions, hidden_states, forward_batch, residual

sglang/srt/models/qwen2_vl.py

@@ -26,7 +26,6 @@ import logging
 from functools import lru_cache, partial
 from typing import Iterable, List, Optional, Tuple, Type, TypedDict
 
-import numpy as np
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
@@ -42,10 +41,11 @@ from sglang.srt.layers.logits_processor import LogitsProcessor
 from sglang.srt.layers.pooler import Pooler, PoolingType
 from sglang.srt.layers.quantization.base_config import QuantizationConfig
 from sglang.srt.layers.vocab_parallel_embedding import ParallelLMHead
-from sglang.srt.managers.multi_modality_padding import (
+from sglang.srt.managers.mm_utils import (
     MultiModalityDataPaddingPatternTokenPairs,
+    general_mm_embed_routine,
 )
-from sglang.srt.managers.schedule_batch import ImageInputs
+from sglang.srt.managers.schedule_batch import MultimodalInputs
 from sglang.srt.model_executor.forward_batch_info import ForwardBatch
 from sglang.srt.model_loader.weight_utils import default_weight_loader
 from sglang.srt.models.qwen2 import Qwen2Model
@@ -351,7 +351,7 @@ class Qwen2VisionTransformer(nn.Module):
 
     @property
     def dtype(self) -> torch.dtype:
-        return self.blocks[0].mlp.fc2.weight.dtype
+        return next(self.parameters()).dtype
 
     @property
     def device(self) -> torch.device:
@@ -359,7 +359,8 @@ class Qwen2VisionTransformer(nn.Module):
 
     def rot_pos_emb(self, grid_thw: torch.Tensor) -> torch.Tensor:
         pos_ids = []
-        for t, h, w in grid_thw:
+        for i in range(grid_thw.size(0)):
+            t, h, w = grid_thw[i].tolist()
             hpos_ids = torch.arange(h).unsqueeze(1).expand(-1, w)
             wpos_ids = torch.arange(w).unsqueeze(0).expand(h, -1)
             hpos_ids = (
@@ -471,18 +472,18 @@ class Qwen2VLForConditionalGeneration(nn.Module):
 
     # Use grid_t * grid_w * grid_h to pad tokens for each image
     # add replaced padding by unique image hash
-    def pad_input_ids(self, input_ids: List[int], image_inputs: ImageInputs):
+    def pad_input_ids(self, input_ids: List[int], multi_modal_inputs: MultimodalInputs):
         # Get all special token IDs
-        im_start_id: int = image_inputs.im_start_id
-        im_end_id: int = image_inputs.im_end_id
+        im_start_id: int = multi_modal_inputs.im_start_id
+        im_end_id: int = multi_modal_inputs.im_end_id
 
         media_token_pairs = [(im_start_id, im_end_id)]
         pattern = MultiModalityDataPaddingPatternTokenPairs(media_token_pairs)
-        return pattern.pad_input_tokens(input_ids, image_inputs)
+        return pattern.pad_input_tokens(input_ids, multi_modal_inputs)
 
-    def _process_image_input(self, image_input: Qwen2VLImageInputs) -> torch.Tensor:
-        pixel_values = image_input["pixel_values"].type(self.visual.dtype)
-        image_embeds = self.visual(pixel_values, grid_thw=image_input["image_grid_thw"])
+    def get_image_feature(self, image_input: MultimodalInputs) -> torch.Tensor:
+        pixel_values = image_input.pixel_values.type(self.visual.dtype)
+        image_embeds = self.visual(pixel_values, grid_thw=image_input.image_grid_thws)
         return image_embeds
 
     def _process_video_input(self, video_input: Qwen2VLVideoInputs) -> torch.Tensor:
@@ -492,6 +493,9 @@ class Qwen2VLForConditionalGeneration(nn.Module):
         )
         return video_embeds
 
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
     def forward(
         self,
         input_ids: torch.Tensor,
@@ -514,67 +518,25 @@ class Qwen2VLForConditionalGeneration(nn.Module):
         if getattr(self.config, "rope_scaling", {}).get("type", None) == "mrope":
             positions = forward_batch.mrope_positions
 
-        image_inputs = None
-        if forward_batch.image_inputs is not None:
-            image_inputs = [
-                img for img in forward_batch.image_inputs if img is not None
-            ]
-
-        if (
+        if not (
             forward_batch.forward_mode.is_decode()
-            or image_inputs is None
-            or len(image_inputs) == 0
+            or not forward_batch.contains_image_inputs()
         ):
-            inputs_embeds = self.model.embed_tokens(input_ids)
-        else:
             if getattr(self.config, "rope_scaling", {}).get("type", None) == "mrope":
                 assert positions.ndim == 2 and positions.size(0) == 3, (
                     "multimodal section rotary embedding requires "
                     f"(3, seq_len) positions, but got {positions.size()}"
                 )
 
-            # Clamp input ids. This is because the input_ids for the image tokens are
-            # filled with the hash values of the image for the prefix matching in the radix attention.
-            # There values are useless because their embeddings will be replaced by vision embeddings anyway.
-            input_ids.clamp_(min=0, max=self.config.vocab_size - 1)
-
-            inputs_embeds = self.model.embed_tokens(input_ids)
-            extend_start_loc_cpu = forward_batch.extend_start_loc.cpu().numpy()
-            prefix_lens_cpu = forward_batch.extend_prefix_lens_cpu
-            for i, image in enumerate(forward_batch.image_inputs):
-                if image is None or image.pixel_values is None:
-                    continue
-                start_idx = extend_start_loc_cpu[i]
-                prefix_len = prefix_lens_cpu[i]
-                pixel_values = image.pixel_values.clone()
-
-                image_grid_thws = torch.tensor(
-                    np.array(image.image_grid_thws), device="cuda"
-                )
-                image_offsets = image.image_offsets
-                image_input = Qwen2VLImageInputs(
-                    pixel_values=pixel_values, image_grid_thw=image_grid_thws
-                )
-                image_embeds = self._process_image_input(image_input)
-
-                image_embeds_offset = 0
-                for idx, image_offset in enumerate(image_offsets):
-                    if image_offset < prefix_len:
-                        continue
-                    num_image_tokens = self.calculate_num_image_tokens(
-                        image_grid_thws[idx]
-                    )
-
-                    left_idx = start_idx + (image_offset - prefix_len + 1)
-                    right_idx = left_idx + num_image_tokens
-                    inputs_embeds[left_idx:right_idx] = image_embeds[
-                        image_embeds_offset : image_embeds_offset + num_image_tokens
-                    ]
-                    image_embeds_offset += num_image_tokens
-            input_ids = None
+        inputs_embeds = general_mm_embed_routine(
+            input_ids=input_ids,
+            forward_batch=forward_batch,
+            embed_tokens=self.get_input_embeddings(),
+            mm_data_embedding_func=self.get_image_feature,
+        )
 
         hidden_states = self.model(
-            input_ids=input_ids,
+            input_ids=None,
             positions=positions,
             forward_batch=forward_batch,
             input_embeds=inputs_embeds,