diff --git a/python/sglang/multimodal_gen/configs/pipeline_configs/base.py b/python/sglang/multimodal_gen/configs/pipeline_configs/base.py
index 80b7758b8..d15506899 100644
--- a/python/sglang/multimodal_gen/configs/pipeline_configs/base.py
+++ b/python/sglang/multimodal_gen/configs/pipeline_configs/base.py
@@ -97,10 +97,6 @@ class STA_Mode(str, Enum):
     NONE = None
 
 
-def preprocess_text(prompt: str) -> str:
-    return prompt
-
-
 def postprocess_text(output: BaseEncoderOutput, _text_inputs) -> torch.tensor:
     raise NotImplementedError
 
@@ -206,8 +202,8 @@ class PipelineConfig:
     def postprocess_image(self, image):
         return image.last_hidden_state
 
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (preprocess_text,)
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (None,)
     )
 
     # get prompt_embeds from encoder output
diff --git a/python/sglang/multimodal_gen/configs/pipeline_configs/flux.py b/python/sglang/multimodal_gen/configs/pipeline_configs/flux.py
index 71d0c0128..9130dfc61 100644
--- a/python/sglang/multimodal_gen/configs/pipeline_configs/flux.py
+++ b/python/sglang/multimodal_gen/configs/pipeline_configs/flux.py
@@ -23,12 +23,10 @@ from sglang.multimodal_gen.configs.models.vaes.flux import Flux2VAEConfig, FluxV
 from sglang.multimodal_gen.configs.pipeline_configs.base import (
     ImagePipelineConfig,
     ModelTaskType,
-    preprocess_text,
     shard_rotary_emb_for_sp,
 )
 from sglang.multimodal_gen.configs.pipeline_configs.hunyuan import (
     clip_postprocess_text,
-    clip_preprocess_text,
 )
 from sglang.multimodal_gen.configs.pipeline_configs.qwen_image import _pack_latents
 from sglang.multimodal_gen.runtime.distributed import get_local_torch_device
@@ -65,8 +63,8 @@ class FluxPipelineConfig(ImagePipelineConfig):
         default_factory=lambda: ("bf16", "bf16")
     )
 
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (clip_preprocess_text, preprocess_text),
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (None, None),
     )
 
     postprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
@@ -650,8 +648,8 @@ class Flux2KleinPipelineConfig(Flux2PipelineConfig):
         default_factory=lambda: (Qwen3TextConfig(),)
     )
 
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (preprocess_text,),
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (None,),
     )
 
     postprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
diff --git a/python/sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py b/python/sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py
index d45dfadb2..abe421c89 100644
--- a/python/sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py
+++ b/python/sglang/multimodal_gen/configs/pipeline_configs/hunyuan.py
@@ -56,10 +56,6 @@ def llama_postprocess_text(outputs: BaseEncoderOutput, _text_inputs) -> torch.te
     return last_hidden_state
 
 
-def clip_preprocess_text(prompt: str) -> str:
-    return prompt
-
-
 def clip_postprocess_text(outputs: BaseEncoderOutput, _text_inputs) -> torch.tensor:
     pooler_output: torch.tensor = outputs.pooler_output
     return pooler_output
@@ -84,8 +80,8 @@ class HunyuanConfig(PipelineConfig):
     text_encoder_configs: tuple[EncoderConfig, ...] = field(
         default_factory=lambda: (LlamaConfig(), CLIPTextConfig())
     )
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (llama_preprocess_text, clip_preprocess_text)
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (llama_preprocess_text, None)
     )
     postprocess_text_funcs: tuple[Callable[[BaseEncoderOutput], torch.tensor], ...] = (
         field(default_factory=lambda: (llama_postprocess_text, clip_postprocess_text))
diff --git a/python/sglang/multimodal_gen/configs/pipeline_configs/ltx_2.py b/python/sglang/multimodal_gen/configs/pipeline_configs/ltx_2.py
index 301423753..54fafce0d 100644
--- a/python/sglang/multimodal_gen/configs/pipeline_configs/ltx_2.py
+++ b/python/sglang/multimodal_gen/configs/pipeline_configs/ltx_2.py
@@ -14,7 +14,6 @@ from sglang.multimodal_gen.configs.models.vaes.ltx_audio import LTXAudioVAEConfi
 from sglang.multimodal_gen.configs.pipeline_configs.base import (
     ModelTaskType,
     PipelineConfig,
-    preprocess_text,
 )
 from sglang.multimodal_gen.runtime.distributed import (
     get_sp_parallel_rank,
@@ -189,8 +188,8 @@ class LTX2PipelineConfig(PipelineConfig):
     text_encoder_precisions: tuple[str, ...] = field(default_factory=lambda: ("bf16",))
     text_encoder_extra_args: list[dict] = field(default_factory=lambda: [{}])
 
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (preprocess_text,)
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (None,)
     )
     postprocess_text_funcs: tuple[
         Callable[[BaseEncoderOutput, dict], torch.Tensor], ...
diff --git a/python/sglang/multimodal_gen/configs/pipeline_configs/sana.py b/python/sglang/multimodal_gen/configs/pipeline_configs/sana.py
index 73f62fa00..d008c0c62 100644
--- a/python/sglang/multimodal_gen/configs/pipeline_configs/sana.py
+++ b/python/sglang/multimodal_gen/configs/pipeline_configs/sana.py
@@ -30,7 +30,6 @@ from sglang.multimodal_gen.configs.models.vaes.sana import SanaVAEConfig
 from sglang.multimodal_gen.configs.pipeline_configs.base import (
     ModelTaskType,
     SpatialImagePipelineConfig,
-    preprocess_text,
 )
 
 
@@ -65,8 +64,8 @@ class SanaPipelineConfig(SpatialImagePipelineConfig):
 
     text_encoder_precisions: tuple[str, ...] = field(default_factory=lambda: ("bf16",))
 
-    preprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
-        default_factory=lambda: (preprocess_text,),
+    preprocess_text_funcs: tuple[Callable[[str], str] | None, ...] = field(
+        default_factory=lambda: (None,),
     )
 
     postprocess_text_funcs: tuple[Callable[[str], str], ...] = field(
diff --git a/python/sglang/multimodal_gen/runtime/pipelines_core/stages/text_encoding.py b/python/sglang/multimodal_gen/runtime/pipelines_core/stages/text_encoding.py
index 7b51cff5f..e1184cbc0 100644
--- a/python/sglang/multimodal_gen/runtime/pipelines_core/stages/text_encoding.py
+++ b/python/sglang/multimodal_gen/runtime/pipelines_core/stages/text_encoding.py
@@ -236,10 +236,12 @@ class TextEncodingStage(PipelineStage):
                 else {}
             )
 
-            processed_text_list: list[str] = []
-            for prompt_str in texts:
-                preprocessed = preprocess_func(prompt_str)
-                processed_text_list.append(preprocessed)
+            if preprocess_func is not None:
+                processed_text_list: list[str] = [
+                    preprocess_func(prompt_str) for prompt_str in texts
+                ]
+            else:
+                processed_text_list = texts
 
             # Prepare tokenizer args
             tok_kwargs = self.prepare_tokenizer_kwargs(