diff --git a/docs/models/supported_models.md b/docs/models/supported_models.md
index e713d7934..269c9261e 100644
--- a/docs/models/supported_models.md
+++ b/docs/models/supported_models.md
@@ -33,6 +33,7 @@ th {
 |`LongCatImageEditPipeline` | LongCat-Image-Edit | `meituan-longcat/LongCat-Image-Edit` |
 |`StableDiffusion3Pipeline` | Stable-Diffusion-3 | `stabilityai/stable-diffusion-3.5-medium` |
 |`Flux2KleinPipeline` | FLUX.2-klein | `black-forest-labs/FLUX.2-klein-4B`, `black-forest-labs/FLUX.2-klein-9B` |
+|`FluxPipeline` | FLUX.1-dev | `black-forest-labs/FLUX.1-dev` |
 |`StableAudioPipeline` | Stable-Audio-Open | `stabilityai/stable-audio-open-1.0` |
 |`Qwen3TTSForConditionalGeneration` | Qwen3-TTS-12Hz-1.7B-CustomVoice | `Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice` |
 |`Qwen3TTSForConditionalGeneration` | Qwen3-TTS-12Hz-1.7B-VoiceDesign | `Qwen/Qwen3-TTS-12Hz-1.7B-VoiceDesign` |
diff --git a/docs/user_guide/diffusion/parallelism_acceleration.md b/docs/user_guide/diffusion/parallelism_acceleration.md
index 45e53294e..1a9f540ad 100644
--- a/docs/user_guide/diffusion/parallelism_acceleration.md
+++ b/docs/user_guide/diffusion/parallelism_acceleration.md
@@ -30,6 +30,7 @@ The following table shows which models are currently supported by parallelism me
 | **Z-Image** | `Tongyi-MAI/Z-Image-Turbo` | ✅ | ✅ | ❌ | ✅ (TP=2 only) |
 | **Stable-Diffusion3.5** | `stabilityai/stable-diffusion-3.5` | ❌ | ❌ | ❌ | ❌ |
 | **FLUX.2-klein** | `black-forest-labs/FLUX.2-klein-4B` | ❌ | ❌ | ❌ | ✅ |
+| **FLUX.1-dev** | `black-forest-labs/FLUX.1-dev` | ❌ | ❌ | ❌ | ✅ |
 
 
 !!! note "TP Limitations for Diffusion Models"
diff --git a/vllm_omni/diffusion/cache/cache_dit_backend.py b/vllm_omni/diffusion/cache/cache_dit_backend.py
index 0c43659c9..ff2c98002 100644
--- a/vllm_omni/diffusion/cache/cache_dit_backend.py
+++ b/vllm_omni/diffusion/cache/cache_dit_backend.py
@@ -791,7 +791,7 @@ def enable_cache_for_bagel(pipeline: Any, cache_config: Any) -> Callable[[int],
 CUSTOM_DIT_ENABLERS.update(
     {
         "WanPipeline": enable_cache_for_wan22,
-        "FluxPipeline": enable_cache_for_flux,
+        # "FluxPipeline": enable_cache_for_flux,
         "LongCatImagePipeline": enable_cache_for_longcat_image,
         "LongCatImageEditPipeline": enable_cache_for_longcat_image,
         "StableDiffusion3Pipeline": enable_cache_for_sd3,
diff --git a/vllm_omni/diffusion/models/flux/__init__.py b/vllm_omni/diffusion/models/flux/__init__.py
new file mode 100644
index 000000000..cad3508a1
--- /dev/null
+++ b/vllm_omni/diffusion/models/flux/__init__.py
@@ -0,0 +1,17 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+"""FLUX.1-dev diffusion model components."""
+
+from vllm_omni.diffusion.models.flux.flux_transformer import (
+    FluxTransformer2DModel,
+)
+from vllm_omni.diffusion.models.flux.pipeline_flux import (
+    FluxPipeline,
+    get_flux_post_process_func,
+)
+
+__all__ = [
+    "FluxPipeline",
+    "FluxTransformer2DModel",
+    "get_flux_post_process_func",
+]
diff --git a/vllm_omni/diffusion/models/flux/flux_transformer.py b/vllm_omni/diffusion/models/flux/flux_transformer.py
new file mode 100644
index 000000000..faf6d08d3
--- /dev/null
+++ b/vllm_omni/diffusion/models/flux/flux_transformer.py
@@ -0,0 +1,638 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+from collections.abc import Iterable
+from typing import Any
+
+import torch
+import torch.nn.functional as F
+from diffusers.models.embeddings import (
+    CombinedTimestepGuidanceTextProjEmbeddings,
+    CombinedTimestepTextProjEmbeddings,
+    get_1d_rotary_pos_embed,
+)
+from diffusers.models.modeling_outputs import Transformer2DModelOutput
+from diffusers.models.normalization import AdaLayerNormContinuous, AdaLayerNormZero, AdaLayerNormZeroSingle
+from diffusers.utils import is_torch_npu_available
+from torch import nn
+from vllm.distributed import get_tensor_model_parallel_world_size, tensor_model_parallel_all_gather
+from vllm.logger import init_logger
+from vllm.model_executor.layers.layernorm import RMSNorm
+from vllm.model_executor.layers.linear import ColumnParallelLinear, QKVParallelLinear, RowParallelLinear
+from vllm.model_executor.model_loader.weight_utils import default_weight_loader
+
+from vllm_omni.diffusion.attention.layer import Attention
+from vllm_omni.diffusion.data import OmniDiffusionConfig
+from vllm_omni.diffusion.layers.rope import RotaryEmbedding
+
+logger = init_logger(__name__)
+
+
+class ColumnParallelApproxGELU(nn.Module):
+    def __init__(self, dim_in: int, dim_out: int, *, approximate: str, bias: bool = True):
+        super().__init__()
+        self.proj = ColumnParallelLinear(
+            dim_in,
+            dim_out,
+            bias=bias,
+            gather_output=False,
+            return_bias=False,
+        )
+        self.approximate = approximate
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        x = self.proj(x)
+        return F.gelu(x, approximate=self.approximate)
+
+
+class FeedForward(nn.Module):
+    def __init__(
+        self,
+        dim: int,
+        dim_out: int | None = None,
+        mult: int = 4,
+        activation_fn: str = "gelu-approximate",
+        inner_dim: int | None = None,
+        bias: bool = True,
+    ) -> None:
+        super().__init__()
+
+        assert activation_fn == "gelu-approximate", "Only gelu-approximate is supported."
+
+        inner_dim = inner_dim or int(dim * mult)
+        dim_out = dim_out or dim
+
+        layers: list[nn.Module] = [
+            ColumnParallelApproxGELU(dim, inner_dim, approximate="tanh", bias=bias),
+            nn.Identity(),  # placeholder for weight loading
+            RowParallelLinear(
+                inner_dim,
+                dim_out,
+                input_is_parallel=True,
+                return_bias=False,
+            ),
+        ]
+
+        self.net = nn.ModuleList(layers)
+
+    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        for module in self.net:
+            hidden_states = module(hidden_states)
+        return hidden_states
+
+
+class FluxAttention(torch.nn.Module):
+    def __init__(
+        self,
+        query_dim: int,
+        heads: int = 8,
+        dim_head: int = 64,
+        dropout: float = 0.0,
+        bias: bool = False,
+        added_kv_proj_dim: int | None = None,
+        added_proj_bias: bool | None = True,
+        out_bias: bool = True,
+        eps: float = 1e-5,
+        out_dim: int = None,
+        context_pre_only: bool | None = None,
+        pre_only: bool = False,
+    ):
+        super().__init__()
+
+        self.head_dim = dim_head
+        self.inner_dim = out_dim if out_dim is not None else dim_head * heads
+        self.query_dim = query_dim
+        self.use_bias = bias
+        self.dropout = dropout
+        self.out_dim = out_dim if out_dim is not None else query_dim
+        self.context_pre_only = context_pre_only
+        self.pre_only = pre_only
+        self.heads = out_dim // dim_head if out_dim is not None else heads
+        self.added_kv_proj_dim = added_kv_proj_dim
+        self.added_proj_bias = added_proj_bias
+
+        self.norm_q = RMSNorm(dim_head, eps=eps)
+        self.norm_k = RMSNorm(dim_head, eps=eps)
+
+        self.to_qkv = QKVParallelLinear(
+            hidden_size=query_dim,
+            head_size=self.head_dim,
+            total_num_heads=self.heads,
+            bias=bias,
+        )
+
+        if not self.pre_only:
+            self.to_out = nn.ModuleList(
+                [
+                    RowParallelLinear(
+                        self.inner_dim,
+                        self.out_dim,
+                        bias=out_bias,
+                        input_is_parallel=True,
+                        return_bias=False,
+                    ),
+                    nn.Dropout(dropout),
+                ]
+            )
+
+        if added_kv_proj_dim is not None:
+            self.norm_added_q = RMSNorm(dim_head, eps=eps)
+            self.norm_added_k = RMSNorm(dim_head, eps=eps)
+
+            self.add_kv_proj = QKVParallelLinear(
+                hidden_size=self.added_kv_proj_dim,
+                head_size=self.head_dim,
+                total_num_heads=self.heads,
+                bias=added_proj_bias,
+            )
+
+            self.to_add_out = RowParallelLinear(
+                self.inner_dim,
+                query_dim,
+                bias=out_bias,
+                input_is_parallel=True,
+                return_bias=False,
+            )
+
+        self.rope = RotaryEmbedding(is_neox_style=False)
+        self.attn = Attention(
+            num_heads=self.to_qkv.num_heads,
+            head_size=self.head_dim,
+            softmax_scale=1.0 / (self.head_dim**0.5),
+            causal=False,
+            num_kv_heads=self.to_qkv.num_kv_heads,
+        )
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        encoder_hidden_states: torch.Tensor | None = None,
+        image_rotary_emb: torch.Tensor | None = None,
+        **kwargs,
+    ) -> torch.Tensor | tuple[torch.Tensor, torch.Tensor]:
+        qkv, _ = self.to_qkv(hidden_states)
+        q_size = self.to_qkv.num_heads * self.head_dim
+        kv_size = self.to_qkv.num_kv_heads * self.head_dim
+        query, key, value = qkv.split([q_size, kv_size, kv_size], dim=-1)
+
+        query = query.unflatten(-1, (self.to_qkv.num_heads, -1))
+        key = key.unflatten(-1, (self.to_qkv.num_kv_heads, -1))
+        value = value.unflatten(-1, (self.to_qkv.num_kv_heads, -1))
+
+        query = self.norm_q(query)
+        key = self.norm_k(key)
+
+        if self.added_kv_proj_dim is not None:
+            encoder_qkv, _ = self.add_kv_proj(encoder_hidden_states)
+            add_q_size = self.add_kv_proj.num_heads * self.head_dim
+            add_kv_size = self.add_kv_proj.num_kv_heads * self.head_dim
+            encoder_query, encoder_key, encoder_value = encoder_qkv.split(
+                [add_q_size, add_kv_size, add_kv_size], dim=-1
+            )
+
+            encoder_query = encoder_query.unflatten(-1, (self.add_kv_proj.num_heads, -1))
+            encoder_key = encoder_key.unflatten(-1, (self.add_kv_proj.num_kv_heads, -1))
+            encoder_value = encoder_value.unflatten(-1, (self.add_kv_proj.num_kv_heads, -1))
+
+            encoder_query = self.norm_added_q(encoder_query)
+            encoder_key = self.norm_added_k(encoder_key)
+
+            query = torch.cat([encoder_query, query], dim=1)
+            key = torch.cat([encoder_key, key], dim=1)
+            value = torch.cat([encoder_value, value], dim=1)
+
+        if image_rotary_emb is not None:
+            cos, sin = image_rotary_emb  # [S, D/2]
+            cos = cos.to(query.dtype)
+            sin = sin.to(query.dtype)
+            query = self.rope(query, cos, sin)
+            key = self.rope(key, cos, sin)
+
+        hidden_states = self.attn(
+            query,
+            key,
+            value,
+        )
+        hidden_states = hidden_states.flatten(2, 3)
+        hidden_states = hidden_states.to(query.dtype)
+
+        if encoder_hidden_states is not None:
+            encoder_hidden_states, hidden_states = hidden_states.split_with_sizes(
+                [encoder_hidden_states.shape[1], hidden_states.shape[1] - encoder_hidden_states.shape[1]], dim=1
+            )
+            hidden_states = self.to_out[0](hidden_states)
+            hidden_states = self.to_out[1](hidden_states)
+            encoder_hidden_states = self.to_add_out(encoder_hidden_states)
+
+            return hidden_states, encoder_hidden_states
+        else:
+            # For single-stream blocks, there's no to_out (RowParallelLinear) to handle the reduction
+            if get_tensor_model_parallel_world_size() > 1:
+                hidden_states = tensor_model_parallel_all_gather(hidden_states, dim=-1)
+            return hidden_states
+
+
+class FluxTransformerBlock(nn.Module):
+    def __init__(
+        self, dim: int, num_attention_heads: int, attention_head_dim: int, qk_norm: str = "rms_norm", eps: float = 1e-6
+    ):
+        super().__init__()
+
+        self.norm1 = AdaLayerNormZero(dim)
+        self.norm1_context = AdaLayerNormZero(dim)
+
+        self.attn = FluxAttention(
+            query_dim=dim,
+            added_kv_proj_dim=dim,
+            dim_head=attention_head_dim,
+            heads=num_attention_heads,
+            out_dim=dim,
+            context_pre_only=False,
+            bias=True,
+            eps=eps,
+        )
+
+        self.norm2 = nn.LayerNorm(dim, elementwise_affine=False, eps=1e-6)
+        self.ff = FeedForward(dim=dim, dim_out=dim)
+
+        self.norm2_context = nn.LayerNorm(dim, elementwise_affine=False, eps=1e-6)
+        self.ff_context = FeedForward(dim=dim, dim_out=dim)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        encoder_hidden_states: torch.Tensor,
+        temb: torch.Tensor,
+        image_rotary_emb: tuple[torch.Tensor, torch.Tensor] | None = None,
+        joint_attention_kwargs: dict[str, Any] | None = None,
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        norm_hidden_states, gate_msa, shift_mlp, scale_mlp, gate_mlp = self.norm1(hidden_states, emb=temb)
+
+        norm_encoder_hidden_states, c_gate_msa, c_shift_mlp, c_scale_mlp, c_gate_mlp = self.norm1_context(
+            encoder_hidden_states, emb=temb
+        )
+        joint_attention_kwargs = joint_attention_kwargs or {}
+
+        # Attention.
+        attention_outputs = self.attn(
+            hidden_states=norm_hidden_states,
+            encoder_hidden_states=norm_encoder_hidden_states,
+            image_rotary_emb=image_rotary_emb,
+            **joint_attention_kwargs,
+        )
+
+        if len(attention_outputs) == 2:
+            attn_output, context_attn_output = attention_outputs
+        elif len(attention_outputs) == 3:
+            attn_output, context_attn_output, ip_attn_output = attention_outputs
+
+        # Process attention outputs for the `hidden_states`.
+        attn_output = gate_msa.unsqueeze(1) * attn_output
+        hidden_states = hidden_states + attn_output
+
+        norm_hidden_states = self.norm2(hidden_states)
+        norm_hidden_states = norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
+
+        ff_output = self.ff(norm_hidden_states)
+        ff_output = gate_mlp.unsqueeze(1) * ff_output
+
+        hidden_states = hidden_states + ff_output
+        if len(attention_outputs) == 3:
+            hidden_states = hidden_states + ip_attn_output
+
+        # Process attention outputs for the `encoder_hidden_states`.
+        context_attn_output = c_gate_msa.unsqueeze(1) * context_attn_output
+        encoder_hidden_states = encoder_hidden_states + context_attn_output
+
+        norm_encoder_hidden_states = self.norm2_context(encoder_hidden_states)
+        norm_encoder_hidden_states = norm_encoder_hidden_states * (1 + c_scale_mlp[:, None]) + c_shift_mlp[:, None]
+
+        context_ff_output = self.ff_context(norm_encoder_hidden_states)
+        encoder_hidden_states = encoder_hidden_states + c_gate_mlp.unsqueeze(1) * context_ff_output
+        if encoder_hidden_states.dtype == torch.float16:
+            encoder_hidden_states = encoder_hidden_states.clip(-65504, 65504)
+
+        return encoder_hidden_states, hidden_states
+
+
+class FluxSingleTransformerBlock(nn.Module):
+    def __init__(self, dim: int, num_attention_heads: int, attention_head_dim: int, mlp_ratio: float = 4.0):
+        super().__init__()
+        self.mlp_hidden_dim = int(dim * mlp_ratio)
+
+        self.norm = AdaLayerNormZeroSingle(dim)
+        self.proj_mlp = nn.Linear(dim, self.mlp_hidden_dim)
+        self.act_mlp = nn.GELU(approximate="tanh")
+        self.proj_out = nn.Linear(dim + self.mlp_hidden_dim, dim)
+
+        self.attn = FluxAttention(
+            query_dim=dim,
+            dim_head=attention_head_dim,
+            heads=num_attention_heads,
+            out_dim=dim,
+            bias=True,
+            eps=1e-6,
+            pre_only=True,
+        )
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        encoder_hidden_states: torch.Tensor,
+        temb: torch.Tensor,
+        image_rotary_emb: tuple[torch.Tensor, torch.Tensor] | None = None,
+        joint_attention_kwargs: dict[str, Any] | None = None,
+    ) -> tuple[torch.Tensor, torch.Tensor]:
+        text_seq_len = encoder_hidden_states.shape[1]
+        hidden_states = torch.cat([encoder_hidden_states, hidden_states], dim=1)
+
+        residual = hidden_states
+        norm_hidden_states, gate = self.norm(hidden_states, emb=temb)
+        mlp_hidden_states = self.act_mlp(self.proj_mlp(norm_hidden_states))
+        joint_attention_kwargs = joint_attention_kwargs or {}
+        attn_output = self.attn(
+            hidden_states=norm_hidden_states,
+            image_rotary_emb=image_rotary_emb,
+            **joint_attention_kwargs,
+        )
+
+        hidden_states = torch.cat([attn_output, mlp_hidden_states], dim=2)
+        gate = gate.unsqueeze(1)
+        hidden_states = gate * self.proj_out(hidden_states)
+        hidden_states = residual + hidden_states
+        if hidden_states.dtype == torch.float16:
+            hidden_states = hidden_states.clip(-65504, 65504)
+
+        encoder_hidden_states, hidden_states = hidden_states[:, :text_seq_len], hidden_states[:, text_seq_len:]
+        return encoder_hidden_states, hidden_states
+
+
+class FluxPosEmbed(nn.Module):
+    # modified from https://github.com/black-forest-labs/flux/blob/c00d7c60b085fce8058b9df845e036090873f2ce/src/flux/modules/layers.py#L11
+    def __init__(self, theta: int, axes_dim: list[int]):
+        super().__init__()
+        self.theta = theta
+        self.axes_dim = axes_dim
+
+    def forward(self, ids: torch.Tensor) -> torch.Tensor:
+        n_axes = ids.shape[-1]
+        cos_out = []
+        sin_out = []
+        pos = ids.float()
+        is_mps = ids.device.type == "mps"
+        is_npu = ids.device.type == "npu"
+        freqs_dtype = torch.float32 if (is_mps or is_npu) else torch.float64
+        for i in range(n_axes):
+            freqs_cis = get_1d_rotary_pos_embed(
+                self.axes_dim[i],
+                pos[:, i],
+                theta=self.theta,
+                use_real=False,
+                freqs_dtype=freqs_dtype,
+            )
+            cos_out.append(freqs_cis.real)
+            sin_out.append(freqs_cis.imag)
+        freqs_cos = torch.cat(cos_out, dim=-1).to(ids.device)
+        freqs_sin = torch.cat(sin_out, dim=-1).to(ids.device)
+        return freqs_cos, freqs_sin
+
+
+class FluxTransformer2DModel(nn.Module):
+    """
+    The Transformer model introduced in Flux.
+
+    Args:
+        od_config (`OmniDiffusionConfig`):
+            The configuration for the model.
+        patch_size (`int`, defaults to `1`):
+            Patch size to turn the input data into small patches.
+        in_channels (`int`, defaults to `64`):
+            The number of channels in the input.
+        out_channels (`int`, *optional*, defaults to `None`):
+            The number of channels in the output. If not specified, it defaults to `in_channels`.
+        num_layers (`int`, defaults to `19`):
+            The number of layers of dual stream DiT blocks to use.
+        num_single_layers (`int`, defaults to `38`):
+            The number of layers of single stream DiT blocks to use.
+        attention_head_dim (`int`, defaults to `128`):
+            The number of dimensions to use for each attention head.
+        num_attention_heads (`int`, defaults to `24`):
+            The number of attention heads to use.
+        joint_attention_dim (`int`, defaults to `4096`):
+            The number of dimensions to use for the joint attention (embedding/channel dimension of
+            `encoder_hidden_states`).
+        pooled_projection_dim (`int`, defaults to `768`):
+            The number of dimensions to use for the pooled projection.
+        guidance_embeds (`bool`, defaults to `False`):
+            Whether to use guidance embeddings for guidance-distilled variant of the model.
+        axes_dims_rope (`Tuple[int]`, defaults to `(16, 56, 56)`):
+            The dimensions to use for the rotary positional embeddings.
+    """
+
+    # the small and frequently-repeated block(s) of a model
+    # -- typically a transformer layer
+    # used for torch compile optimizations
+    _repeated_blocks = ["FluxTransformerBlock"]
+
+    def __init__(
+        self,
+        od_config: OmniDiffusionConfig,
+        patch_size: int = 1,
+        in_channels: int = 64,
+        out_channels: int = None,
+        num_layers: int = 19,
+        num_single_layers: int = 38,
+        attention_head_dim: int = 128,
+        num_attention_heads: int = 24,
+        joint_attention_dim: int = 4096,
+        pooled_projection_dim: int = 768,
+        guidance_embeds: bool = True,
+        axes_dims_rope: tuple[int, int, int] = (16, 56, 56),
+    ):
+        super().__init__()
+        model_config = od_config.tf_model_config
+        num_layers = model_config.num_layers
+        self.parallel_config = od_config.parallel_config
+        self.in_channels = in_channels
+        self.out_channels = out_channels or in_channels
+        self.inner_dim = num_attention_heads * attention_head_dim
+        self.guidance_embeds = guidance_embeds
+
+        self.pos_embed = FluxPosEmbed(theta=10000, axes_dim=axes_dims_rope)
+        text_time_guidance_cls = (
+            CombinedTimestepGuidanceTextProjEmbeddings if guidance_embeds else CombinedTimestepTextProjEmbeddings
+        )
+        self.time_text_embed = text_time_guidance_cls(
+            embedding_dim=self.inner_dim, pooled_projection_dim=pooled_projection_dim
+        )
+
+        self.context_embedder = nn.Linear(joint_attention_dim, self.inner_dim)
+        self.x_embedder = nn.Linear(in_channels, self.inner_dim)
+
+        self.transformer_blocks = nn.ModuleList(
+            [
+                FluxTransformerBlock(
+                    dim=self.inner_dim,
+                    num_attention_heads=num_attention_heads,
+                    attention_head_dim=attention_head_dim,
+                )
+                for _ in range(num_layers)
+            ]
+        )
+
+        self.single_transformer_blocks = nn.ModuleList(
+            [
+                FluxSingleTransformerBlock(
+                    dim=self.inner_dim,
+                    num_attention_heads=num_attention_heads,
+                    attention_head_dim=attention_head_dim,
+                )
+                for _ in range(num_single_layers)
+            ]
+        )
+
+        self.norm_out = AdaLayerNormContinuous(self.inner_dim, self.inner_dim, elementwise_affine=False, eps=1e-6)
+        self.proj_out = nn.Linear(self.inner_dim, patch_size * patch_size * self.out_channels, bias=True)
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        encoder_hidden_states: torch.Tensor = None,
+        pooled_projections: torch.Tensor = None,
+        timestep: torch.LongTensor = None,
+        img_ids: torch.Tensor = None,
+        txt_ids: torch.Tensor = None,
+        guidance: torch.Tensor | None = None,
+        joint_attention_kwargs: dict[str, Any] | None = None,
+        return_dict: bool = True,
+    ) -> torch.Tensor | Transformer2DModelOutput:
+        """
+        The [`FluxTransformer2DModel`] forward method.
+
+        Args:
+            hidden_states (`torch.Tensor` of shape `(batch_size, image_sequence_length, in_channels)`):
+                Input `hidden_states`.
+            encoder_hidden_states (`torch.Tensor` of shape `(batch_size, text_sequence_length, joint_attention_dim)`):
+                Conditional embeddings (embeddings computed from the input conditions such as prompts) to use.
+            pooled_projections (`torch.Tensor` of shape `(batch_size, projection_dim)`): Embeddings projected
+                from the embeddings of input conditions.
+            timestep ( `torch.LongTensor`):
+                Used to indicate denoising step.
+            img_ids: (`torch.Tensor`):
+                The position ids for image tokens.
+            txt_ids (`torch.Tensor`):
+                The position ids for text tokens.
+            guidance (`torch.Tensor`):
+                Guidance embeddings for guidance-distilled variant of the model.
+            joint_attention_kwargs (`dict`, *optional*):
+                A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
+                `self.processor` in
+                [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+            return_dict (`bool`, *optional*, defaults to `True`):
+                Whether or not to return a [`~models.transformer_2d.Transformer2DModelOutput`] instead of a plain
+                tuple.
+
+        Returns:
+            If `return_dict` is True, an [`~models.transformer_2d.Transformer2DModelOutput`] is returned, otherwise a
+            `tuple` where the first element is the sample tensor.
+        """
+
+        hidden_states = self.x_embedder(hidden_states)
+        timestep = timestep.to(device=hidden_states.device, dtype=hidden_states.dtype) * 1000
+
+        if guidance is not None:
+            guidance = guidance.to(device=hidden_states.device, dtype=hidden_states.dtype) * 1000
+
+        temb = (
+            self.time_text_embed(timestep, pooled_projections)
+            if guidance is None
+            else self.time_text_embed(timestep, guidance, pooled_projections)
+        )
+        encoder_hidden_states = self.context_embedder(encoder_hidden_states)
+
+        if txt_ids.ndim == 3:
+            logger.warning(
+                "Passing `txt_ids` 3d torch.Tensor is deprecated."
+                "Please remove the batch dimension and pass it as a 2d torch Tensor"
+            )
+            txt_ids = txt_ids[0]
+        if img_ids.ndim == 3:
+            logger.warning(
+                "Passing `img_ids` 3d torch.Tensor is deprecated."
+                "Please remove the batch dimension and pass it as a 2d torch Tensor"
+            )
+            img_ids = img_ids[0]
+
+        ids = torch.cat((txt_ids, img_ids), dim=0)
+        if is_torch_npu_available():
+            freqs_cos, freqs_sin = self.pos_embed(ids.cpu())
+            image_rotary_emb = (freqs_cos.npu(), freqs_sin.npu())
+        else:
+            image_rotary_emb = self.pos_embed(ids)
+
+        for index_block, block in enumerate(self.transformer_blocks):
+            encoder_hidden_states, hidden_states = block(
+                hidden_states=hidden_states,
+                encoder_hidden_states=encoder_hidden_states,
+                temb=temb,
+                image_rotary_emb=image_rotary_emb,
+                joint_attention_kwargs=joint_attention_kwargs,
+            )
+
+        for index_block, block in enumerate(self.single_transformer_blocks):
+            encoder_hidden_states, hidden_states = block(
+                hidden_states=hidden_states,
+                encoder_hidden_states=encoder_hidden_states,
+                temb=temb,
+                image_rotary_emb=image_rotary_emb,
+                joint_attention_kwargs=joint_attention_kwargs,
+            )
+
+        hidden_states = self.norm_out(hidden_states, temb)
+        output = self.proj_out(hidden_states)
+
+        if not return_dict:
+            return (output,)
+
+        return Transformer2DModelOutput(sample=output)
+
+    def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]) -> set[str]:
+        stacked_params_mapping = [
+            # (param_name, shard_name, shard_id)
+            # self-attn
+            (".to_qkv", ".to_q", "q"),
+            (".to_qkv", ".to_k", "k"),
+            (".to_qkv", ".to_v", "v"),
+            # cross-attn
+            (".add_kv_proj", ".add_q_proj", "q"),
+            (".add_kv_proj", ".add_k_proj", "k"),
+            (".add_kv_proj", ".add_v_proj", "v"),
+        ]
+
+        params_dict = dict(self.named_parameters())
+
+        # we need to load the buffers for beta and eps (XIELU)
+        for name, buffer in self.named_buffers():
+            if name.endswith(".beta") or name.endswith(".eps"):
+                params_dict[name] = buffer
+
+        loaded_params: set[str] = set()
+        for name, loaded_weight in weights:
+            original_name = name
+            lookup_name = name
+            for param_name, weight_name, shard_id in stacked_params_mapping:
+                if weight_name not in original_name:
+                    continue
+                lookup_name = original_name.replace(weight_name, param_name)
+                param = params_dict[lookup_name]
+                weight_loader = param.weight_loader
+                weight_loader(param, loaded_weight, shard_id)
+                break
+            else:
+                if lookup_name not in params_dict and ".to_out.0." in lookup_name:
+                    lookup_name = lookup_name.replace(".to_out.0.", ".to_out.")
+                param = params_dict[lookup_name]
+                weight_loader = getattr(param, "weight_loader", default_weight_loader)
+                weight_loader(param, loaded_weight)
+            loaded_params.add(original_name)
+            loaded_params.add(lookup_name)
+        return loaded_params
diff --git a/vllm_omni/diffusion/models/flux/pipeline_flux.py b/vllm_omni/diffusion/models/flux/pipeline_flux.py
new file mode 100644
index 000000000..01e0bfba9
--- /dev/null
+++ b/vllm_omni/diffusion/models/flux/pipeline_flux.py
@@ -0,0 +1,712 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+import inspect
+import json
+import logging
+import os
+from collections.abc import Iterable
+from typing import Any
+
+import numpy as np
+import torch
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.loaders import TextualInversionLoaderMixin
+from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL
+from diffusers.schedulers.scheduling_flow_match_euler_discrete import (
+    FlowMatchEulerDiscreteScheduler,
+)
+from diffusers.utils.torch_utils import randn_tensor
+from torch import nn
+from transformers import CLIPTextModel, CLIPTokenizer, T5EncoderModel, T5TokenizerFast
+from vllm.model_executor.models.utils import AutoWeightsLoader
+
+from vllm_omni.diffusion.data import DiffusionOutput, OmniDiffusionConfig
+from vllm_omni.diffusion.distributed.utils import get_local_device
+from vllm_omni.diffusion.model_loader.diffusers_loader import DiffusersPipelineLoader
+from vllm_omni.diffusion.models.flux import FluxTransformer2DModel
+from vllm_omni.diffusion.request import OmniDiffusionRequest
+from vllm_omni.model_executor.model_loader.weight_utils import download_weights_from_hf_specific
+
+logger = logging.getLogger(__name__)
+
+
+def get_flux_post_process_func(
+    od_config: OmniDiffusionConfig,
+):
+    if od_config.output_type == "latent":
+        return lambda x: x
+    model_name = od_config.model
+    if os.path.exists(model_name):
+        model_path = model_name
+    else:
+        model_path = download_weights_from_hf_specific(model_name, None, ["*"])
+
+    vae_config_path = os.path.join(model_path, "vae/config.json")
+    with open(vae_config_path) as f:
+        vae_config = json.load(f)
+        vae_scale_factor = 2 ** (len(vae_config["block_out_channels"]) - 1) if "block_out_channels" in vae_config else 8
+
+    image_processor = VaeImageProcessor(vae_scale_factor=vae_scale_factor * 2)
+
+    def post_process_func(images: torch.Tensor):
+        return image_processor.postprocess(images)
+
+    return post_process_func
+
+
+def calculate_shift(
+    image_seq_len,
+    base_seq_len: int = 256,
+    max_seq_len: int = 4096,
+    base_shift: float = 0.5,
+    max_shift: float = 1.15,
+):
+    m = (max_shift - base_shift) / (max_seq_len - base_seq_len)
+    b = base_shift - m * base_seq_len
+    mu = image_seq_len * m + b
+    return mu
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.retrieve_timesteps
+def retrieve_timesteps(
+    scheduler,
+    num_inference_steps: int | None = None,
+    device: str | torch.device | None = None,
+    timesteps: list[int] | None = None,
+    sigmas: list[float] | None = None,
+    **kwargs,
+) -> tuple[torch.Tensor, int]:
+    r"""
+    Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles
+    custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`.
+
+    Args:
+        scheduler (`SchedulerMixin`):
+            The scheduler to get timesteps from.
+        num_inference_steps (`int`):
+            The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+            must be `None`.
+        device (`str` or `torch.device`, *optional*):
+            The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
+        timesteps (`List[int]`, *optional*):
+            Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+            `num_inference_steps` and `sigmas` must be `None`.
+        sigmas (`List[float]`, *optional*):
+            Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+            `num_inference_steps` and `timesteps` must be `None`.
+
+    Returns:
+        `tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
+        second element is the number of inference steps.
+    """
+    if timesteps is not None and sigmas is not None:
+        raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
+    if timesteps is not None:
+        accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accepts_timesteps:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" timestep schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
+    elif sigmas is not None:
+        accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+        if not accept_sigmas:
+            raise ValueError(
+                f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+                f" sigmas schedules. Please check whether you are using the correct scheduler."
+            )
+        scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+        num_inference_steps = len(timesteps)
+    else:
+        scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
+        timesteps = scheduler.timesteps
+    return timesteps, num_inference_steps
+
+
+class FluxPipeline(
+    nn.Module,
+):
+    def __init__(
+        self,
+        *,
+        od_config: OmniDiffusionConfig,
+        prefix: str = "",
+    ):
+        super().__init__()
+        self.od_config = od_config
+        self.weights_sources = [
+            DiffusersPipelineLoader.ComponentSource(
+                model_or_path=od_config.model,
+                subfolder="transformer",
+                revision=None,
+                prefix="transformer.",
+                fall_back_to_pt=True,
+            )
+        ]
+
+        self.device = get_local_device()
+        model = od_config.model
+        # Check if model is a local path
+        local_files_only = os.path.exists(model)
+
+        self.scheduler = FlowMatchEulerDiscreteScheduler.from_pretrained(
+            model, subfolder="scheduler", local_files_only=local_files_only
+        )
+        self.text_encoder = CLIPTextModel.from_pretrained(
+            model, subfolder="text_encoder", local_files_only=local_files_only
+        )
+        self.text_encoder_2 = T5EncoderModel.from_pretrained(
+            model, subfolder="text_encoder_2", local_files_only=local_files_only
+        )
+        self.vae = AutoencoderKL.from_pretrained(model, subfolder="vae", local_files_only=local_files_only).to(
+            self.device
+        )
+        self.transformer = FluxTransformer2DModel(od_config=od_config)
+
+        self.tokenizer = CLIPTokenizer.from_pretrained(model, subfolder="tokenizer", local_files_only=local_files_only)
+        self.tokenizer_2 = T5TokenizerFast.from_pretrained(
+            model, subfolder="tokenizer_2", local_files_only=local_files_only
+        )
+
+        self.stage = None
+
+        self.vae_scale_factor = 2 ** (len(self.vae.config.block_out_channels) - 1) if getattr(self, "vae", None) else 8
+        # Flux latents are turned into 2x2 patches and packed.
+        # This means the latent width and height has to be divisible
+        # by the patch size. So the vae scale factor is multiplied by the patch size to account for this
+        # self.image_processor = VaeImageProcessor(vae_scale_factor=self.vae_scale_factor * 2)
+        self.tokenizer_max_length = (
+            self.tokenizer.model_max_length if hasattr(self, "tokenizer") and self.tokenizer is not None else 77
+        )
+        self.default_sample_size = 128
+
+    def check_inputs(
+        self,
+        prompt,
+        prompt_2,
+        height,
+        width,
+        negative_prompt=None,
+        negative_prompt_2=None,
+        prompt_embeds=None,
+        negative_prompt_embeds=None,
+        pooled_prompt_embeds=None,
+        negative_pooled_prompt_embeds=None,
+        callback_on_step_end_tensor_inputs=None,
+        max_sequence_length=None,
+    ):
+        if height % (self.vae_scale_factor * 2) != 0 or width % (self.vae_scale_factor * 2) != 0:
+            logger.warning(
+                f"`height` and `width` have to be divisible by {self.vae_scale_factor * 2} "
+                f"but are {height} and {width}. Dimensions will be resized accordingly"
+            )
+
+        if prompt is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt_2 is not None and prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `prompt_2`: {prompt_2} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+                " only forward one of the two."
+            )
+        elif prompt is None and prompt_embeds is None:
+            raise ValueError(
+                "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+            )
+        elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+            raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+        elif prompt_2 is not None and (not isinstance(prompt_2, str) and not isinstance(prompt_2, list)):
+            raise ValueError(f"`prompt_2` has to be of type `str` or `list` but is {type(prompt_2)}")
+
+        if negative_prompt is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+        elif negative_prompt_2 is not None and negative_prompt_embeds is not None:
+            raise ValueError(
+                f"Cannot forward both `negative_prompt_2`: {negative_prompt_2} and `negative_prompt_embeds`:"
+                f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+            )
+
+        if prompt_embeds is not None and pooled_prompt_embeds is None:
+            raise ValueError(
+                "If `prompt_embeds` are provided, `pooled_prompt_embeds` also have to be passed. "
+                "Make sure to generate `pooled_prompt_embeds` from the same text encoder "
+                "that was used to generate `prompt_embeds`."
+            )
+        if negative_prompt_embeds is not None and negative_pooled_prompt_embeds is None:
+            raise ValueError(
+                "If `negative_prompt_embeds` are provided, `negative_pooled_prompt_embeds` also have to be passed. "
+                "Make sure to generate `negative_pooled_prompt_embeds` from the same text encoder "
+                "that was used to generate `negative_prompt_embeds`."
+            )
+
+        if max_sequence_length is not None and max_sequence_length > 512:
+            raise ValueError(f"`max_sequence_length` cannot be greater than 512 but is {max_sequence_length}")
+
+    def _get_t5_prompt_embeds(
+        self,
+        prompt: str | list[str] = None,
+        num_images_per_prompt: int = 1,
+        max_sequence_length: int = 512,
+        dtype: torch.dtype | None = None,
+    ):
+        prompt = [prompt] if isinstance(prompt, str) else prompt
+        batch_size = len(prompt)
+
+        if isinstance(self, TextualInversionLoaderMixin):
+            prompt = self.maybe_convert_prompt(prompt, self.tokenizer_2)
+
+        text_inputs = self.tokenizer_2(
+            prompt,
+            padding="max_length",
+            max_length=max_sequence_length,
+            truncation=True,
+            return_length=False,
+            return_overflowing_tokens=False,
+            return_tensors="pt",
+        )
+        text_input_ids = text_inputs.input_ids
+        untruncated_ids = self.tokenizer_2(prompt, padding="longest", return_tensors="pt").input_ids
+
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer_2.batch_decode(untruncated_ids[:, max_sequence_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because `max_sequence_length` is set to "
+                f" {max_sequence_length} tokens: {removed_text}"
+            )
+
+        prompt_embeds = self.text_encoder_2(text_input_ids.to(self.device), output_hidden_states=False)[0]
+
+        dtype = self.text_encoder_2.dtype
+        prompt_embeds = prompt_embeds.to(dtype=dtype, device=self.device)
+
+        _, seq_len, _ = prompt_embeds.shape
+
+        # duplicate text embeddings and attention mask for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt, 1)
+        prompt_embeds = prompt_embeds.view(batch_size * num_images_per_prompt, seq_len, -1)
+
+        return prompt_embeds
+
+    def _get_clip_prompt_embeds(
+        self,
+        prompt: str | list[str],
+        num_images_per_prompt: int = 1,
+    ):
+        prompt = [prompt] if isinstance(prompt, str) else prompt
+        batch_size = len(prompt)
+
+        if isinstance(self, TextualInversionLoaderMixin):
+            prompt = self.maybe_convert_prompt(prompt, self.tokenizer)
+
+        text_inputs = self.tokenizer(
+            prompt,
+            padding="max_length",
+            max_length=self.tokenizer_max_length,
+            truncation=True,
+            return_overflowing_tokens=False,
+            return_length=False,
+            return_tensors="pt",
+        )
+
+        text_input_ids = text_inputs.input_ids
+        untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+        if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(text_input_ids, untruncated_ids):
+            removed_text = self.tokenizer.batch_decode(untruncated_ids[:, self.tokenizer_max_length - 1 : -1])
+            logger.warning(
+                "The following part of your input was truncated because CLIP can only handle sequences up to"
+                f" {self.tokenizer_max_length} tokens: {removed_text}"
+            )
+        prompt_embeds = self.text_encoder(text_input_ids.to(self.device), output_hidden_states=False)
+
+        # Use pooled output of CLIPTextModel
+        prompt_embeds = prompt_embeds.pooler_output
+        prompt_embeds = prompt_embeds.to(dtype=self.text_encoder.dtype, device=self.device)
+
+        # duplicate text embeddings for each generation per prompt, using mps friendly method
+        prompt_embeds = prompt_embeds.repeat(1, num_images_per_prompt)
+        prompt_embeds = prompt_embeds.view(batch_size * num_images_per_prompt, -1)
+
+        return prompt_embeds
+
+    def encode_prompt(
+        self,
+        prompt: str | list[str],
+        prompt_2: str | list[str],
+        num_images_per_prompt: int = 1,
+        prompt_embeds: torch.FloatTensor | None = None,
+        pooled_prompt_embeds: torch.FloatTensor | None = None,
+        max_sequence_length: int = 512,
+    ):
+        r"""
+
+        Args:
+            prompt (`str` or `List[str]`, *optional*):
+                prompt to be encoded
+            prompt_2 (`str` or `List[str]`, *optional*):
+                The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is
+                used in all text-encoders
+            num_images_per_prompt (`int`):
+                number of images that should be generated per prompt
+            prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
+                provided, text embeddings will be generated from `prompt` input argument.
+            pooled_prompt_embeds (`torch.FloatTensor`, *optional*):
+                Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting.
+                If not provided, pooled text embeddings will be generated from `prompt` input argument.
+        """
+
+        prompt = [prompt] if isinstance(prompt, str) else prompt
+
+        if prompt_embeds is None:
+            prompt_2 = prompt_2 or prompt
+            prompt_2 = [prompt_2] if isinstance(prompt_2, str) else prompt_2
+
+            # We only use the pooled prompt output from the CLIPTextModel
+            pooled_prompt_embeds = self._get_clip_prompt_embeds(
+                prompt=prompt,
+                num_images_per_prompt=num_images_per_prompt,
+            )
+            prompt_embeds = self._get_t5_prompt_embeds(
+                prompt=prompt_2,
+                num_images_per_prompt=num_images_per_prompt,
+                max_sequence_length=max_sequence_length,
+            )
+
+        dtype = self.text_encoder.dtype if self.text_encoder is not None else self.transformer.dtype
+        text_ids = torch.zeros(prompt_embeds.shape[1], 3).to(device=self.device, dtype=dtype)
+
+        return prompt_embeds, pooled_prompt_embeds, text_ids
+
+    @staticmethod
+    def _prepare_latent_image_ids(batch_size, height, width, device, dtype):
+        latent_image_ids = torch.zeros(height, width, 3)
+        latent_image_ids[..., 1] = latent_image_ids[..., 1] + torch.arange(height)[:, None]
+        latent_image_ids[..., 2] = latent_image_ids[..., 2] + torch.arange(width)[None, :]
+
+        latent_image_id_height, latent_image_id_width, latent_image_id_channels = latent_image_ids.shape
+
+        latent_image_ids = latent_image_ids.reshape(
+            latent_image_id_height * latent_image_id_width, latent_image_id_channels
+        )
+
+        return latent_image_ids.to(device=device, dtype=dtype)
+
+    @staticmethod
+    def _pack_latents(latents, batch_size, num_channels_latents, height, width):
+        latents = latents.view(batch_size, num_channels_latents, height // 2, 2, width // 2, 2)
+        latents = latents.permute(0, 2, 4, 1, 3, 5)
+        latents = latents.reshape(batch_size, (height // 2) * (width // 2), num_channels_latents * 4)
+
+        return latents
+
+    @staticmethod
+    def _unpack_latents(latents, height, width, vae_scale_factor):
+        batch_size, num_patches, channels = latents.shape
+
+        # VAE applies 8x compression on images but we must also account for packing which requires
+        # latent height and width to be divisible by 2.
+        height = 2 * (int(height) // (vae_scale_factor * 2))
+        width = 2 * (int(width) // (vae_scale_factor * 2))
+
+        latents = latents.view(batch_size, height // 2, width // 2, channels // 4, 2, 2)
+        latents = latents.permute(0, 3, 1, 4, 2, 5)
+
+        latents = latents.reshape(batch_size, channels // (2 * 2), height, width)
+
+        return latents
+
+    def prepare_latents(
+        self,
+        batch_size,
+        num_channels_latents,
+        height,
+        width,
+        dtype,
+        device,
+        generator,
+        latents=None,
+    ):
+        # VAE applies 8x compression on images but we must also account for packing which requires
+        # latent height and width to be divisible by 2.
+        height = 2 * (int(height) // (self.vae_scale_factor * 2))
+        width = 2 * (int(width) // (self.vae_scale_factor * 2))
+
+        shape = (batch_size, num_channels_latents, height, width)
+
+        if latents is not None:
+            latent_image_ids = self._prepare_latent_image_ids(batch_size, height // 2, width // 2, device, dtype)
+            return latents.to(device=device, dtype=dtype), latent_image_ids
+
+        if isinstance(generator, list) and len(generator) != batch_size:
+            raise ValueError(
+                f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+                f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+            )
+
+        latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+        latents = self._pack_latents(latents, batch_size, num_channels_latents, height, width)
+
+        latent_image_ids = self._prepare_latent_image_ids(batch_size, height // 2, width // 2, device, dtype)
+
+        return latents, latent_image_ids
+
+    def prepare_timesteps(self, num_inference_steps, sigmas, image_seq_len):
+        sigmas = np.linspace(1.0, 1 / num_inference_steps, num_inference_steps) if sigmas is None else sigmas
+        mu = calculate_shift(
+            image_seq_len,
+            self.scheduler.config.get("base_image_seq_len", 256),
+            self.scheduler.config.get("max_image_seq_len", 4096),
+            self.scheduler.config.get("base_shift", 0.5),
+            self.scheduler.config.get("max_shift", 1.15),
+        )
+        timesteps, num_inference_steps = retrieve_timesteps(
+            self.scheduler,
+            num_inference_steps,
+            sigmas=sigmas,
+            mu=mu,
+        )
+        return timesteps, num_inference_steps
+
+    @property
+    def guidance_scale(self):
+        return self._guidance_scale
+
+    @property
+    def joint_attention_kwargs(self):
+        return self._joint_attention_kwargs
+
+    @property
+    def num_timesteps(self):
+        return self._num_timesteps
+
+    @property
+    def current_timestep(self):
+        return self._current_timestep
+
+    @property
+    def interrupt(self):
+        return self._interrupt
+
+    def diffuse(
+        self,
+        prompt_embeds,
+        pooled_prompt_embeds,
+        negative_prompt_embeds,
+        negative_pooled_prompt_embeds,
+        latents,
+        latent_image_ids,
+        text_ids,
+        negative_text_ids,
+        timesteps,
+        do_true_cfg,
+        guidance,
+        true_cfg_scale,
+    ):
+        """Diffusion loop with optional image conditioning."""
+        self.scheduler.set_begin_index(0)
+        for i, t in enumerate(timesteps):
+            if self.interrupt:
+                continue
+
+            self._current_timestep = t
+            # broadcast to batch dimension and place on same device/dtype as latents
+            timestep = t.expand(latents.shape[0]).to(device=latents.device, dtype=latents.dtype)
+
+            self.transformer.do_true_cfg = do_true_cfg  # used in teacache hook
+            # Forward pass for positive prompt (or unconditional if no CFG)
+            noise_pred = self.transformer(
+                hidden_states=latents,
+                timestep=timestep / 1000,
+                guidance=guidance,
+                pooled_projections=pooled_prompt_embeds,
+                encoder_hidden_states=prompt_embeds,
+                txt_ids=text_ids,
+                img_ids=latent_image_ids,
+                joint_attention_kwargs=self.joint_attention_kwargs,
+                return_dict=False,
+            )[0]
+
+            # Forward pass for negative prompt (CFG)
+            if do_true_cfg:
+                neg_noise_pred = self.transformer(
+                    hidden_states=latents,
+                    timestep=timestep / 1000,
+                    guidance=guidance,
+                    pooled_projections=negative_pooled_prompt_embeds,
+                    encoder_hidden_states=negative_prompt_embeds,
+                    txt_ids=negative_text_ids,
+                    img_ids=latent_image_ids,
+                    joint_attention_kwargs=self.joint_attention_kwargs,
+                    return_dict=False,
+                )[0]
+                noise_pred = neg_noise_pred + true_cfg_scale * (noise_pred - neg_noise_pred)
+            # compute the previous noisy sample x_t -> x_t-1
+            latents = self.scheduler.step(noise_pred, t, latents, return_dict=False)[0]
+        return latents
+
+    def forward(
+        self,
+        req: OmniDiffusionRequest,
+        prompt: str | list[str] = None,
+        prompt_2: str | list[str] = None,
+        negative_prompt: str | list[str] = None,
+        negative_prompt_2: str | list[str] = None,
+        true_cfg_scale: float = 1.0,
+        height: int | None = None,
+        width: int | None = None,
+        num_inference_steps: int = 28,
+        sigmas: list[float] | None = None,
+        guidance_scale: float = 3.5,
+        num_images_per_prompt: int = 1,
+        generator: torch.Generator | list[torch.Generator] | None = None,
+        latents: torch.FloatTensor | None = None,
+        prompt_embeds: torch.FloatTensor | None = None,
+        pooled_prompt_embeds: torch.FloatTensor | None = None,
+        negative_prompt_embeds: torch.FloatTensor | None = None,
+        negative_pooled_prompt_embeds: torch.FloatTensor | None = None,
+        output_type: str | None = "pil",
+        return_dict: bool = True,
+        joint_attention_kwargs: dict[str, Any] | None = None,
+        callback_on_step_end_tensor_inputs: list[str] = ["latents"],
+        max_sequence_length: int = 512,
+    ):
+        """Forward pass for flux."""
+        prompt = req.prompt if req.prompt is not None else prompt
+        negative_prompt = req.negative_prompt if req.negative_prompt is not None else negative_prompt
+        height = req.height or self.default_sample_size * self.vae_scale_factor
+        width = req.width or self.default_sample_size * self.vae_scale_factor
+        sigmas = req.sigmas or sigmas
+        num_inference_steps = req.num_inference_steps or num_inference_steps
+        generator = req.generator or generator
+        req_num_outputs = getattr(req, "num_outputs_per_prompt", None)
+        if req_num_outputs and req_num_outputs > 0:
+            num_images_per_prompt = req_num_outputs
+
+        # 1. Check inputs. Raise error if not correct
+        self.check_inputs(
+            prompt,
+            prompt_2,
+            height,
+            width,
+            negative_prompt=negative_prompt,
+            negative_prompt_2=negative_prompt_2,
+            prompt_embeds=prompt_embeds,
+            negative_prompt_embeds=negative_prompt_embeds,
+            pooled_prompt_embeds=pooled_prompt_embeds,
+            negative_pooled_prompt_embeds=negative_pooled_prompt_embeds,
+            callback_on_step_end_tensor_inputs=callback_on_step_end_tensor_inputs,
+            max_sequence_length=max_sequence_length,
+        )
+
+        self._guidance_scale = guidance_scale
+        self._joint_attention_kwargs = joint_attention_kwargs
+        self._current_timestep = None
+        self._interrupt = False
+
+        # 2. Define call parameters
+        if prompt is not None and isinstance(prompt, str):
+            batch_size = 1
+        elif prompt is not None and isinstance(prompt, list):
+            batch_size = len(prompt)
+        else:
+            batch_size = prompt_embeds.shape[0]
+
+        has_neg_prompt = negative_prompt is not None or (
+            negative_prompt_embeds is not None and negative_pooled_prompt_embeds is not None
+        )
+        do_true_cfg = true_cfg_scale > 1 and has_neg_prompt
+
+        (
+            prompt_embeds,
+            pooled_prompt_embeds,
+            text_ids,
+        ) = self.encode_prompt(
+            prompt=prompt,
+            prompt_2=prompt_2,
+            prompt_embeds=prompt_embeds,
+            pooled_prompt_embeds=pooled_prompt_embeds,
+            num_images_per_prompt=num_images_per_prompt,
+            max_sequence_length=max_sequence_length,
+        )
+
+        negative_text_ids = None
+        if do_true_cfg:
+            (
+                negative_prompt_embeds,
+                negative_pooled_prompt_embeds,
+                negative_text_ids,
+            ) = self.encode_prompt(
+                prompt=negative_prompt,
+                prompt_2=negative_prompt_2,
+                prompt_embeds=negative_prompt_embeds,
+                pooled_prompt_embeds=negative_pooled_prompt_embeds,
+                num_images_per_prompt=num_images_per_prompt,
+                max_sequence_length=max_sequence_length,
+            )
+
+        # 4. Prepare latent variables
+        num_channels_latents = self.transformer.in_channels // 4
+        latents, latent_image_ids = self.prepare_latents(
+            batch_size * num_images_per_prompt,
+            num_channels_latents,
+            height,
+            width,
+            prompt_embeds.dtype,
+            self.device,
+            generator,
+            latents,
+        )
+
+        # 5. Prepare timesteps
+        timesteps, num_inference_steps = self.prepare_timesteps(num_inference_steps, sigmas, latents.shape[1])
+        self._num_timesteps = len(timesteps)
+
+        # handle guidance
+        if self.transformer.guidance_embeds:
+            guidance = torch.full([1], guidance_scale, dtype=torch.float32)
+            guidance = guidance.expand(latents.shape[0])
+        else:
+            guidance = None
+
+        if self.joint_attention_kwargs is None:
+            self._joint_attention_kwargs = {}
+
+        latents = self.diffuse(
+            prompt_embeds,
+            pooled_prompt_embeds,
+            negative_prompt_embeds,
+            negative_pooled_prompt_embeds,
+            latents,
+            latent_image_ids,
+            text_ids,
+            negative_text_ids,
+            timesteps,
+            do_true_cfg,
+            guidance,
+            true_cfg_scale,
+        )
+
+        self._current_timestep = None
+
+        if output_type == "latent":
+            image = latents
+        else:
+            latents = self._unpack_latents(latents, height, width, self.vae_scale_factor)
+            latents = (latents / self.vae.config.scaling_factor) + self.vae.config.shift_factor
+            image = self.vae.decode(latents, return_dict=False)[0]
+
+        return DiffusionOutput(output=image)
+
+    def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]) -> set[str]:
+        loader = AutoWeightsLoader(self)
+        return loader.load_weights(weights)
diff --git a/vllm_omni/diffusion/registry.py b/vllm_omni/diffusion/registry.py
index d4429aff6..024352479 100644
--- a/vllm_omni/diffusion/registry.py
+++ b/vllm_omni/diffusion/registry.py
@@ -90,6 +90,11 @@ _DIFFUSION_MODELS = {
         "pipeline_flux2_klein",
         "Flux2KleinPipeline",
     ),
+    "FluxPipeline": (
+        "flux",
+        "pipeline_flux",
+        "FluxPipeline",
+    ),
 }
 
 
@@ -227,6 +232,7 @@ _DIFFUSION_POST_PROCESS_FUNCS = {
     "LongCatImageEditPipeline": "get_longcat_image_post_process_func",
     "StableDiffusion3Pipeline": "get_sd3_image_post_process_func",
     "Flux2KleinPipeline": "get_flux2_klein_post_process_func",
+    "FluxPipeline": "get_flux_post_process_func",
 }
 
 _DIFFUSION_PRE_PROCESS_FUNCS = {