From 208966d929edb4fdb810d16e3f5c92f12bc351c8 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 11:20:34 +0800
Subject: [PATCH 1/6] ovis2.5 supports PP

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 vllm/model_executor/models/ovis2_5.py      |  36 ++--
 vllm/model_executor/models/siglip2navit.py | 223 ++++++++++++++-------
 2 files changed, 171 insertions(+), 88 deletions(-)

diff --git a/vllm/model_executor/models/ovis2_5.py b/vllm/model_executor/models/ovis2_5.py
index aa4ea3dd48f6..58a14072443c 100644
--- a/vllm/model_executor/models/ovis2_5.py
+++ b/vllm/model_executor/models/ovis2_5.py
@@ -30,7 +30,7 @@
 from vllm.sequence import IntermediateTensors
 from vllm.transformers_utils.processors.ovis2_5 import Ovis2_5Processor
 
-from .interfaces import MultiModalEmbeddings, SupportsMultiModal
+from .interfaces import MultiModalEmbeddings, SupportsMultiModal, SupportsPP
 
 IMAGE_TOKEN = "<image>"
 VIDEO_TOKEN = "<video>"
@@ -70,6 +70,7 @@ def __init__(
         visual_vocab_size: int,
         quant_config: Optional[QuantizationConfig] = None,
         prefix: str = "",
+        use_data_parallel: bool = False,
     ):
         super().__init__()
         self.config = config
@@ -77,6 +78,7 @@ def __init__(
             config=config,
             quant_config=quant_config,
             prefix=f"{prefix}.vit",
+            use_data_parallel=use_data_parallel,
         )
         # reserved tokens for INDICATOR_IDS
         head_dim = visual_vocab_size - len(INDICATOR_IDS)
@@ -93,31 +95,33 @@ def _init_backbone(
         config: PretrainedConfig,
         quant_config: Optional[QuantizationConfig] = None,
         prefix: str = "",
+        use_data_parallel: bool = False,
     ):
         model_type = config.model_type
         if model_type == "siglip2_navit":
-            return Siglip2NavitModel(config=config, )
+            return Siglip2NavitModel(config=config,
+                                     quant_config=quant_config,
+                                     prefix=prefix,
+                                     use_data_parallel=use_data_parallel)
         raise ValueError(
             f"Unsupported visual tokenizer model_type: {model_type}")
 
     @property
-    def dtype(self):
+    def dtype(self) -> torch.dtype:
         return next(self.head.parameters()).dtype
 
     @property
-    def device(self):
+    def device(self) -> torch.device:
         return next(self.head.parameters()).device
 
-    def tokenize(self, logits):
+    def tokenize(self, logits: torch.Tensor) -> torch.Tensor:
         tokens = torch.softmax(logits, dim=-1,
                                dtype=torch.float32).to(logits.dtype)
         return tokens
 
-    def encode(self, pixel_values, grid_thws):
-        features = self.vit(pixel_values,
-                            grid_thws,
-                            output_hidden_states=True,
-                            return_dict=True)
+    def encode(self, pixel_values: torch.Tensor,
+               grid_thws: torch.Tensor) -> torch.Tensor:
+        features = self.vit(pixel_values, grid_thws)
         # refer to qwen2.5-vl patchmerger
         seq_len, _ = features.shape
         features = features.reshape(seq_len // (self.config.hidden_stride**2),
@@ -125,7 +129,8 @@ def encode(self, pixel_values, grid_thws):
 
         return features
 
-    def forward(self, pixel_values, grid_thws) -> torch.Tensor:
+    def forward(self, pixel_values: torch.Tensor,
+                grid_thws: torch.Tensor) -> torch.Tensor:
         features = self.encode(pixel_values, grid_thws)
         logits = self.head(features)
         tokens = self.tokenize(logits)
@@ -395,7 +400,7 @@ def get_replacement_ovis(item_idx, modality: str):
 @MULTIMODAL_REGISTRY.register_processor(Ovis2_5MultiModalProcessor,
                                         info=Ovis2_5ProcessingInfo,
                                         dummy_inputs=Ovis2_5DummyInputsBuilder)
-class Ovis2_5(nn.Module, SupportsMultiModal):
+class Ovis2_5(nn.Module, SupportsMultiModal, SupportsPP):
 
     def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         super().__init__()
@@ -421,9 +426,8 @@ def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
         text_model_type = self.config.get_text_config().model_type
         self.image_pad_token_id = IMAGE_PAD_TOKEN_ID_MAP[text_model_type]
 
-        # TODO(Isotr0py): PP support
-        # self.make_empty_intermediate_tensors = (
-        #    self.language_model.make_empty_intermediate_tensors)
+        self.make_empty_intermediate_tensors = (
+            self.get_language_model().make_empty_intermediate_tensors)
 
     def _parse_and_validate_visual_input(
             self, is_video,
@@ -567,4 +571,4 @@ def load_weights(self, weights: Iterable[tuple[str,
         return loader.load_weights(weights)
 
     def get_language_model(self) -> torch.nn.Module:
-        return self.llm
\ No newline at end of file
+        return self.llm
diff --git a/vllm/model_executor/models/siglip2navit.py b/vllm/model_executor/models/siglip2navit.py
index 10093f92a5c1..5241c1df55df 100644
--- a/vllm/model_executor/models/siglip2navit.py
+++ b/vllm/model_executor/models/siglip2navit.py
@@ -3,16 +3,23 @@
 """Implementation of SiglipVisionModel intended to be only used
 within a vision language model."""
 
-from typing import Optional, Union
+from collections.abc import Iterable
+from typing import Optional
 
 import torch
 from einops import rearrange, repeat
 from torch import nn
 from torch.nn import functional as F
-from transformers.activations import ACT2FN
+from transformers import Siglip2VisionConfig
 from transformers.configuration_utils import PretrainedConfig
-from transformers.modeling_outputs import BaseModelOutputWithNoAttention
 
+from vllm.config import QuantizationConfig
+from vllm.model_executor.layers.activation import get_act_fn
+from vllm.model_executor.layers.linear import (ColumnParallelLinear,
+                                               LinearBase, QKVParallelLinear,
+                                               ReplicatedLinear,
+                                               RowParallelLinear)
+from vllm.model_executor.model_loader.weight_utils import default_weight_loader
 from vllm.platforms import _Backend
 
 from .vision import get_vit_attn_backend
@@ -48,10 +55,11 @@ def __init__(self, config: PretrainedConfig):
 
         # siglip2 naflex
         if self.num_patches > 0:
-            self.patch_embedding = nn.Linear(
-                in_features=config.num_channels * self.patch_size *
+            self.patch_embedding = ReplicatedLinear(
+                input_size=config.num_channels * self.patch_size *
                 self.patch_size,
-                out_features=self.embed_dim,
+                output_size=self.embed_dim,
+                return_bias=False,
             )
             if self.preserve_original_pe:
                 self.position_embedding_size = int(self.num_patches**0.5)
@@ -89,7 +97,7 @@ def forward(self,
 
         # Apply patch embeddings to already patchified pixel values
         target_dtype = self.patch_embedding.weight.dtype
-        if isinstance(self.patch_embedding, nn.Linear):
+        if isinstance(self.patch_embedding, LinearBase):
             patch_embeds = self.patch_embedding(
                 pixel_values.to(dtype=target_dtype))
         elif isinstance(self.patch_embedding, nn.Conv2d):
@@ -184,7 +192,13 @@ def apply_rotary_pos_emb(
 class Siglip2Attention(nn.Module):
     """Multi-headed attention from 'Attention Is All You Need' paper"""
 
-    def __init__(self, config):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
         self.config = config
         self.embed_dim = config.hidden_size
@@ -199,10 +213,21 @@ def __init__(self, config):
         self.dropout = config.attention_dropout
         self.is_causal = False
 
-        self.k_proj = nn.Linear(self.embed_dim, self.embed_dim)
-        self.v_proj = nn.Linear(self.embed_dim, self.embed_dim)
-        self.q_proj = nn.Linear(self.embed_dim, self.embed_dim)
-        self.out_proj = nn.Linear(self.embed_dim, self.embed_dim)
+        # TODO(Isotr0py): Enable data parallel after we support
+        # disabling TP on parallel linear layer
+        self.qkv_proj = QKVParallelLinear(
+            hidden_size=self.embed_dim,
+            head_size=self.head_dim,
+            total_num_heads=self.num_heads,
+            quant_config=quant_config,
+            prefix=f"{prefix}.qkv_proj",
+        )
+        self.out_proj = RowParallelLinear(
+            input_size=self.embed_dim,
+            output_size=self.embed_dim,
+            quant_config=quant_config,
+            prefix=f"{prefix}.out_proj",
+        )
 
         self.use_rope = config.use_rope
 
@@ -228,9 +253,8 @@ def forward(
 
         seq_length, embed_dim = hidden_states.shape
 
-        queries = self.q_proj(hidden_states)
-        keys = self.k_proj(hidden_states)
-        values = self.v_proj(hidden_states)
+        qkv_states, _ = self.qkv_proj(hidden_states)
+        queries, keys, values = qkv_states.chunk(3, dim=-1)
 
         queries = queries.view(seq_length, self.num_heads, self.head_dim)
         keys = keys.view(seq_length, self.num_heads, self.head_dim)
@@ -286,12 +310,30 @@ def forward(
 
 class Siglip2MLP(nn.Module):
 
-    def __init__(self, config):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
         self.config = config
-        self.activation_fn = ACT2FN[config.hidden_act]
-        self.fc1 = nn.Linear(config.hidden_size, config.intermediate_size)
-        self.fc2 = nn.Linear(config.intermediate_size, config.hidden_size)
+        self.activation_fn = get_act_fn(config.hidden_act)
+        # TODO(Isotr0py): Enable data parallel after we support
+        # disabling TP on parallel linear layer
+        self.fc1 = ColumnParallelLinear(
+            config.hidden_size,
+            config.intermediate_size,
+            quant_config=quant_config,
+            prefix=f"{prefix}.fc1",
+        )
+        self.fc2 = RowParallelLinear(
+            config.intermediate_size,
+            config.hidden_size,
+            quant_config=quant_config,
+            prefix=f"{prefix}.fc2",
+        )
 
     def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
         hidden_states = self.fc1(hidden_states)
@@ -302,15 +344,27 @@ def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
 
 class Siglip2EncoderLayer(nn.Module):
 
-    def __init__(self, config: PretrainedConfig):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
         self.embed_dim = config.hidden_size
         self.layer_norm1 = nn.LayerNorm(self.embed_dim,
                                         eps=config.layer_norm_eps)
-        self.self_attn = Siglip2Attention(config)
+        self.self_attn = Siglip2Attention(config,
+                                          quant_config=quant_config,
+                                          prefix=f"{prefix}.self_attn",
+                                          use_data_parallel=use_data_parallel)
         self.layer_norm2 = nn.LayerNorm(self.embed_dim,
                                         eps=config.layer_norm_eps)
-        self.mlp = Siglip2MLP(config)
+        self.mlp = Siglip2MLP(config,
+                              quant_config=quant_config,
+                              prefix=f"{prefix}.mlp",
+                              use_data_parallel=use_data_parallel)
 
     def forward(self, hidden_states: torch.Tensor, cu_seqlens: torch.Tensor,
                 position_embeddings: torch.Tensor) -> tuple[torch.FloatTensor]:
@@ -347,14 +401,22 @@ class Siglip2Encoder(nn.Module):
         config: PretrainedConfig
     """
 
-    def __init__(self, config: PretrainedConfig):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
         self.config = config
         self.layers = nn.ModuleList([
-            Siglip2EncoderLayer(config)
-            for _ in range(config.num_hidden_layers)
+            Siglip2EncoderLayer(config,
+                                quant_config=quant_config,
+                                prefix=f"{prefix}.layers.{idx}",
+                                use_data_parallel=use_data_parallel)
+            for idx in range(config.num_hidden_layers)
         ])
-        self.gradient_checkpointing = False
 
         self.rotary_pos_emb = VisionRotaryEmbedding(
             config.hidden_size // config.num_attention_heads // 2)
@@ -445,12 +507,10 @@ def get_window_index(self, grid_thw):
 
         return window_index, cu_window_seqlens
 
-    # Ignore copy
     def forward(
         self,
-        inputs_embeds,
+        inputs_embeds: torch.Tensor,
         grid_thws: torch.Tensor,
-        output_hidden_states: bool = False,
     ) -> tuple[torch.Tensor, Optional[tuple[torch.Tensor, ...]]]:
         r"""
         Args:
@@ -506,7 +566,6 @@ def forward(
         cu_seqlens = F.pad(cu_seqlens, (1, 0), value=0)
 
         reverse_indices = torch.argsort(window_index)
-        encoder_states = () if output_hidden_states else None
 
         hidden_states = inputs_embeds
         for index, block in enumerate(self.layers):
@@ -517,45 +576,41 @@ def forward(
                 cu_seqlens_tmp = cu_window_seqlens
             hidden_states = block(hidden_states, cu_seqlens_tmp,
                                   position_embeddings)
-            if output_hidden_states:
-                hidden_states_ = hidden_states.reshape(
-                    seq_len // self.spatial_merge_unit,
-                    self.spatial_merge_unit, -1)
-                encoder_states += (hidden_states_[reverse_indices, :].reshape(
-                    seq_len, -1), )
+
         # tokens = self.post_trunk_norm(tokens)
         hidden_states = hidden_states.reshape(
             seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1)
         hidden_states = hidden_states[reverse_indices, :].reshape(seq_len, -1)
 
-        return hidden_states, encoder_states
+        return hidden_states
 
 
 class Siglip2VisionTransformer(nn.Module):
 
-    def __init__(self, config: PretrainedConfig):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
         self.config = config
         embed_dim = config.hidden_size
 
         self.embeddings = Siglip2VisionEmbeddings(config)
-        self.encoder = Siglip2Encoder(config)
+        self.encoder = Siglip2Encoder(config,
+                                      quant_config=quant_config,
+                                      prefix=f"{prefix}.encoder",
+                                      use_data_parallel=use_data_parallel)
         self.post_layernorm = nn.LayerNorm(embed_dim,
                                            eps=config.layer_norm_eps)
-        self._use_flash_attention_2 = \
-            (config._attn_implementation == "flash_attention_2")
 
     def forward(
         self,
         pixel_values: torch.FloatTensor,
         grid_thws: torch.LongTensor,
-        output_hidden_states: Optional[bool] = True,
-        return_dict: Optional[bool] = True,
-    ) -> Union[
-            tuple[torch.Tensor],
-            tuple[torch.Tensor, tuple[torch.Tensor, ...]],
-            BaseModelOutputWithNoAttention,
-    ]:
+    ) -> torch.Tensor:
         r"""
         spatial_shapes (`torch.LongTensor` of shape `(batch_size, 2)`):
             Tensor containing the spatial dimensions (height, width)
@@ -563,45 +618,69 @@ def forward(
         """
         hidden_states = self.embeddings(pixel_values, grid_thws)
 
-        last_hidden_state, hidden_states = self.encoder(
-            hidden_states, grid_thws, output_hidden_states)
+        last_hidden_state = self.encoder(hidden_states, grid_thws)
         last_hidden_state = self.post_layernorm(last_hidden_state)
 
-        if not return_dict:
-            output = (last_hidden_state, )
-            output += (hidden_states, ) if output_hidden_states else ()
-            return output
-
         return last_hidden_state
 
 
 class Siglip2NavitModel(torch.nn.Module):
 
-    def __init__(self, config: PretrainedConfig):
+    def __init__(
+        self,
+        config: Siglip2VisionConfig,
+        quant_config: Optional[QuantizationConfig] = None,
+        prefix: str = "",
+        use_data_parallel: bool = False,
+    ):
         super().__init__()
 
-        self.vision_model = Siglip2VisionTransformer(config)
+        self.vision_model = Siglip2VisionTransformer(
+            config,
+            quant_config=quant_config,
+            prefix=f"{prefix}.vision_model",
+            use_data_parallel=use_data_parallel)
 
     def forward(
         self,
         pixel_values: torch.FloatTensor,
         grid_thws: torch.LongTensor,
-        output_hidden_states: Optional[bool] = None,
-        return_dict: Optional[bool] = None,
-    ) -> Union[
-            tuple[torch.Tensor],
-            tuple[torch.Tensor, tuple[torch.Tensor, ...]],
-            BaseModelOutputWithNoAttention,
-    ]:
-
-        if output_hidden_states is None:
-            output_hidden_states = self.config.output_hidden_states
-        if return_dict is None:
-            return_dict = self.config.use_return_dict
-
+    ) -> torch.Tensor:
         return self.vision_model(
             pixel_values=pixel_values,
             grid_thws=grid_thws,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
         )
+
+    def load_weights(self, weights: Iterable[tuple[str,
+                                                   torch.Tensor]]) -> set[str]:
+        stacked_params_mapping = [
+            # (param_name, shard_name, shard_id)
+            ("qkv_proj", "q_proj", "q"),
+            ("qkv_proj", "k_proj", "k"),
+            ("qkv_proj", "v_proj", "v"),
+        ]
+        params_dict = dict(self.named_parameters())
+        loaded_params: set[str] = set()
+
+        for name, loaded_weight in weights:
+            # post_layernorm is optional in SiglipVisionModel
+            if (name.startswith("trunk.post_trunk_norm")
+                    and self.trunk.post_trunk_norm is None):
+                continue
+
+            for (param_name, weight_name, shard_id) in stacked_params_mapping:
+                if weight_name not in name:
+                    continue
+                name = name.replace(weight_name, param_name)
+
+                param = params_dict[name]
+                weight_loader = param.weight_loader
+                weight_loader(param, loaded_weight, shard_id)
+                break
+            else:
+                param = params_dict[name]
+                weight_loader = getattr(param, "weight_loader",
+                                        default_weight_loader)
+                weight_loader(param, loaded_weight)
+            loaded_params.add(name)
+        return loaded_params

From b6d62cf4ede5f7feb6b2fc099d39201b34161301 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 11:43:36 +0800
Subject: [PATCH 2/6] fix

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 vllm/model_executor/models/siglip2navit.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/vllm/model_executor/models/siglip2navit.py b/vllm/model_executor/models/siglip2navit.py
index 5241c1df55df..ab215a27b5d5 100644
--- a/vllm/model_executor/models/siglip2navit.py
+++ b/vllm/model_executor/models/siglip2navit.py
@@ -304,7 +304,7 @@ def forward(
                 outputs.append(output_i)
 
             attn_output = torch.cat(outputs, dim=0)
-        attn_output = self.out_proj(attn_output)
+        attn_output, _ = self.out_proj(attn_output)
         return attn_output
 
 
@@ -511,7 +511,7 @@ def forward(
         self,
         inputs_embeds: torch.Tensor,
         grid_thws: torch.Tensor,
-    ) -> tuple[torch.Tensor, Optional[tuple[torch.Tensor, ...]]]:
+    ) -> torch.Tensor:
         r"""
         Args:
             inputs_embeds (`torch.FloatTensor` of shape

From 6619ec80c2f0d42a84f8fe99242aa490b17d0d26 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 12:41:32 +0800
Subject: [PATCH 3/6] fix tp

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 vllm/model_executor/models/siglip2navit.py | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/vllm/model_executor/models/siglip2navit.py b/vllm/model_executor/models/siglip2navit.py
index ab215a27b5d5..5018fb2d647a 100644
--- a/vllm/model_executor/models/siglip2navit.py
+++ b/vllm/model_executor/models/siglip2navit.py
@@ -14,6 +14,7 @@
 from transformers.configuration_utils import PretrainedConfig
 
 from vllm.config import QuantizationConfig
+from vllm.distributed import divide, get_tensor_model_parallel_world_size
 from vllm.model_executor.layers.activation import get_act_fn
 from vllm.model_executor.layers.linear import (ColumnParallelLinear,
                                                LinearBase, QKVParallelLinear,
@@ -229,6 +230,9 @@ def __init__(
             prefix=f"{prefix}.out_proj",
         )
 
+        self.tp_size = (1 if use_data_parallel else
+                        get_tensor_model_parallel_world_size())
+        self.num_heads_per_partition = divide(self.num_heads, self.tp_size)
         self.use_rope = config.use_rope
 
         # Detect attention implementation.
@@ -256,9 +260,12 @@ def forward(
         qkv_states, _ = self.qkv_proj(hidden_states)
         queries, keys, values = qkv_states.chunk(3, dim=-1)
 
-        queries = queries.view(seq_length, self.num_heads, self.head_dim)
-        keys = keys.view(seq_length, self.num_heads, self.head_dim)
-        values = values.view(seq_length, self.num_heads, self.head_dim)
+        queries = queries.view(seq_length, self.num_heads_per_partition,
+                               self.head_dim)
+        keys = keys.view(seq_length, self.num_heads_per_partition,
+                         self.head_dim)
+        values = values.view(seq_length, self.num_heads_per_partition,
+                             self.head_dim)
 
         if self.use_rope:
             cos, sin = position_embeddings
@@ -300,7 +307,8 @@ def forward(
                                                           v_i,
                                                           dropout_p=0.0)
                 # (1, num_heads, seq_len, head_dim) -> (seq_len, embed_dim)
-                output_i = output_i.transpose(1, 2).reshape(-1, self.embed_dim)
+                output_i = output_i.transpose(1, 2).reshape(
+                    end_idx - start_idx, -1)
                 outputs.append(output_i)
 
             attn_output = torch.cat(outputs, dim=0)
@@ -336,9 +344,9 @@ def __init__(
         )
 
     def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        hidden_states = self.fc1(hidden_states)
+        hidden_states, _ = self.fc1(hidden_states)
         hidden_states = self.activation_fn(hidden_states)
-        hidden_states = self.fc2(hidden_states)
+        hidden_states, _ = self.fc2(hidden_states)
         return hidden_states
 
 

From 79f77a5f42a461a319297725b244c627d70a6b05 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 22:13:59 +0800
Subject: [PATCH 4/6] use repo's PR with sdpa fallback

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 tests/distributed/test_pipeline_parallel.py | 1 +
 tests/models/registry.py                    | 3 +--
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/tests/distributed/test_pipeline_parallel.py b/tests/distributed/test_pipeline_parallel.py
index 12dd7c422263..28150d768237 100644
--- a/tests/distributed/test_pipeline_parallel.py
+++ b/tests/distributed/test_pipeline_parallel.py
@@ -233,6 +233,7 @@ def iter_params(self, model_id: str):
     "openbmb/MiniCPM-Llama3-V-2_5": PPTestSettings.fast(),
     "allenai/Molmo-7B-D-0924": PPTestSettings.fast(),
     "AIDC-AI/Ovis2-1B": PPTestSettings.fast(),
+    "AIDC-AI/Ovis2.5-2B": PPTestSettings.fast(),
     "microsoft/Phi-3.5-vision-instruct": PPTestSettings.fast(),
     "mistralai/Pixtral-12B-2409": PPTestSettings.fast(load_format="dummy"),
     "Qwen/Qwen-VL-Chat": PPTestSettings.fast(),
diff --git a/tests/models/registry.py b/tests/models/registry.py
index a6d5c305f799..b99998227dcc 100644
--- a/tests/models/registry.py
+++ b/tests/models/registry.py
@@ -466,8 +466,7 @@ def check_available_online(
                                     "1.6-gemma": "AIDC-AI/Ovis1.6-Gemma2-9B"}),  # noqa: E501
     "Ovis2_5": _HfExamplesInfo("AIDC-AI/Ovis2.5-2B",
                                trust_remote_code=True,
-                               max_transformers_version="4.53",
-                               transformers_version_reason="HF model is not compatible"),  # noqa: E501
+                               revision="refs/pr/5"),
     "PaliGemmaForConditionalGeneration": _HfExamplesInfo("google/paligemma-3b-mix-224",  # noqa: E501
                                                          extras={"v2": "google/paligemma2-3b-ft-docci-448"}),  # noqa: E501
     "Phi3VForCausalLM": _HfExamplesInfo("microsoft/Phi-3-vision-128k-instruct",

From 7438de8af31e6308cd0db53d107f785e859d0090 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 23:00:13 +0800
Subject: [PATCH 5/6] HF sdpa fallback

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 tests/models/multimodal/generation/test_common.py | 6 +-----
 tests/models/registry.py                          | 3 +--
 2 files changed, 2 insertions(+), 7 deletions(-)

diff --git a/tests/models/multimodal/generation/test_common.py b/tests/models/multimodal/generation/test_common.py
index ea5de9d9f5c5..96208f8eda62 100644
--- a/tests/models/multimodal/generation/test_common.py
+++ b/tests/models/multimodal/generation/test_common.py
@@ -11,7 +11,6 @@
 import pytest
 from transformers import (AutoModel, AutoModelForImageTextToText,
                           AutoModelForTextToWaveform, AutoModelForVision2Seq)
-from transformers.utils import is_flash_attn_2_available
 
 from vllm.platforms import current_platform
 from vllm.utils import identity
@@ -637,10 +636,7 @@
         dtype="half",
         num_logprobs=10,
         patch_hf_runner=model_utils.ovis2_5_patch_hf_runner,
-        marks=[pytest.mark.skipif(
-            not is_flash_attn_2_available(),
-            reason="HF model needs `flash_attn` installed"
-        )],
+        hf_model_kwargs={"revision": "refs/pr/5"},
     ),
     "phi3v": VLMTestInfo(
         models=["microsoft/Phi-3.5-vision-instruct"],
diff --git a/tests/models/registry.py b/tests/models/registry.py
index b99998227dcc..3e142e7dfdf7 100644
--- a/tests/models/registry.py
+++ b/tests/models/registry.py
@@ -465,8 +465,7 @@ def check_available_online(
                             extras={"1.6-llama": "AIDC-AI/Ovis1.6-Llama3.2-3B",
                                     "1.6-gemma": "AIDC-AI/Ovis1.6-Gemma2-9B"}),  # noqa: E501
     "Ovis2_5": _HfExamplesInfo("AIDC-AI/Ovis2.5-2B",
-                               trust_remote_code=True,
-                               revision="refs/pr/5"),
+                               trust_remote_code=True),
     "PaliGemmaForConditionalGeneration": _HfExamplesInfo("google/paligemma-3b-mix-224",  # noqa: E501
                                                          extras={"v2": "google/paligemma2-3b-ft-docci-448"}),  # noqa: E501
     "Phi3VForCausalLM": _HfExamplesInfo("microsoft/Phi-3-vision-128k-instruct",

From ac4cf0f3edee0388f35374bbf90d4023ae0ebad0 Mon Sep 17 00:00:00 2001
From: Isotr0py <mozf@mail2.sysu.edu.cn>
Date: Fri, 22 Aug 2025 23:22:20 +0800
Subject: [PATCH 6/6] gemini

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
---
 vllm/model_executor/models/siglip2navit.py | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/vllm/model_executor/models/siglip2navit.py b/vllm/model_executor/models/siglip2navit.py
index 5018fb2d647a..c6244fb3b3e6 100644
--- a/vllm/model_executor/models/siglip2navit.py
+++ b/vllm/model_executor/models/siglip2navit.py
@@ -585,7 +585,6 @@ def forward(
             hidden_states = block(hidden_states, cu_seqlens_tmp,
                                   position_embeddings)
 
-        # tokens = self.post_trunk_norm(tokens)
         hidden_states = hidden_states.reshape(
             seq_len // self.spatial_merge_unit, self.spatial_merge_unit, -1)
         hidden_states = hidden_states[reverse_indices, :].reshape(seq_len, -1)
@@ -671,11 +670,6 @@ def load_weights(self, weights: Iterable[tuple[str,
         loaded_params: set[str] = set()
 
         for name, loaded_weight in weights:
-            # post_layernorm is optional in SiglipVisionModel
-            if (name.startswith("trunk.post_trunk_norm")
-                    and self.trunk.post_trunk_norm is None):
-                continue
-
             for (param_name, weight_name, shard_id) in stacked_params_mapping:
                 if weight_name not in name:
                     continue