cosmos2_5

class Cosmos25PatchEmbed(nn.Module):
    """
    COSMOS 2.5 patch embedding - converts video (B, C, T, H, W) to patches (B, T', H', W', D).
    Uses linear projection after rearranging patches.
    """
def __init__(
    self,
    in_channels: int,
    out_channels: int,
    patch_size: tuple[int, int, int],
) -> None:
    super().__init__()
    self.patch_size = patch_size
    self.dim = in_channels * patch_size[0] * patch_size[1] * patch_size[2]

    self.proj = nn.Linear(self.dim, out_channels, bias=False)

def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, C, T, H, W)
    Returns:
        (B, T', H', W', D) where T'=T//pt, H'=H//ph, W'=W//pw
    """
    batch_size, num_channels, num_frames, height, width = hidden_states.shape
    p_t, p_h, p_w = self.patch_size

    # Rearrange: b c (t pt) (h ph) (w pw) -> b t h w (c pt ph pw)
    hidden_states = hidden_states.reshape(batch_size, num_channels, num_frames // p_t, p_t, height // p_h, p_h,
                                          width // p_w, p_w)
    hidden_states = hidden_states.permute(0, 2, 4, 6, 1, 3, 5, 7)
    hidden_states = hidden_states.flatten(4, 7)  # Flatten patch dimensions

    # Project to model dimension — cast to weight dtype to handle fp32/bf16 mismatch
    # between training (fp32) and inference (bf16) contexts
    hidden_states = self.proj(hidden_states.to(self.proj.weight.dtype))
    return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25PatchEmbed.forward
forward(hidden_states: torch.Tensor) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, C, T, H, W)

Returns:
    (B, T', H', W', D) where T'=T//pt, H'=H//ph, W'=W//pw
Source
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, C, T, H, W)
    Returns:
        (B, T', H', W', D) where T'=T//pt, H'=H//ph, W'=W//pw
    """
    batch_size, num_channels, num_frames, height, width = hidden_states.shape
    p_t, p_h, p_w = self.patch_size
# Rearrange: b c (t pt) (h ph) (w pw) -&gt; b t h w (c pt ph pw)
hidden_states = hidden_states.reshape(batch_size, num_channels, num_frames // p_t, p_t, height // p_h, p_h,
                                      width // p_w, p_w)
hidden_states = hidden_states.permute(0, 2, 4, 6, 1, 3, 5, 7)
hidden_states = hidden_states.flatten(4, 7)  # Flatten patch dimensions

# Project to model dimension — cast to weight dtype to handle fp32/bf16 mismatch
# between training (fp32) and inference (bf16) contexts
hidden_states = self.proj(hidden_states.to(self.proj.weight.dtype))
return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25TimestepEmbedding
Cosmos25TimestepEmbedding(in_features: int, out_features: int, use_adaln_lora: bool = True, adaln_lora_dim: int = 256)
COSMOS 2.5 timestep embedding with AdaLN-LoRA support.
Generates both standard embedding and AdaLN-LoRA parameters.
Source
class Cosmos25TimestepEmbedding(nn.Module):
    """
    COSMOS 2.5 timestep embedding with AdaLN-LoRA support.
    Generates both standard embedding and AdaLN-LoRA parameters.
    """
def __init__(
    self,
    in_features: int,
    out_features: int,
    use_adaln_lora: bool = True,
    adaln_lora_dim: int = 256,
) -&gt; None:
    super().__init__()
    self.use_adaln_lora = use_adaln_lora

    self.linear_1 = nn.Linear(in_features, out_features, bias=False)
    self.activation = nn.SiLU()

    if use_adaln_lora:
        self.linear_2 = nn.Linear(out_features, 3 * out_features, bias=False)
    else:
        self.linear_2 = nn.Linear(out_features, out_features, bias=False)

def forward(self, sample: torch.Tensor) -&gt; tuple&#91;torch.Tensor, torch.Tensor | None&#93;:
    &quot;&quot;&quot;
    Returns:
        emb: Standard embedding (B, T, D)
        adaln_lora: AdaLN-LoRA parameters (B, T, 3D) or None
    &quot;&quot;&quot;
    emb = self.linear_1(sample)
    emb = self.activation(emb)
    emb = self.linear_2(emb)

    if self.use_adaln_lora:
        adaln_lora = emb  # (B, T, 3D)
        emb_standard = sample  # Use input as standard embedding
    else:
        emb_standard = emb
        adaln_lora = None

    return emb_standard, adaln_lora</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25TimestepEmbedding.forward
forward(sample: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor | None]
Returns

emb: torch.Tensor
Standard embedding (B, T, D)
adaln_lora: torch.Tensor | None
AdaLN-LoRA parameters (B, T, 3D) or None

Source
def forward(self, sample: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor | None]:
    """
    Returns:
        emb: Standard embedding (B, T, D)
        adaln_lora: AdaLN-LoRA parameters (B, T, 3D) or None
    """
    emb = self.linear_1(sample)
    emb = self.activation(emb)
    emb = self.linear_2(emb)
if self.use_adaln_lora:
    adaln_lora = emb  # (B, T, 3D)
    emb_standard = sample  # Use input as standard embedding
else:
    emb_standard = emb
    adaln_lora = None

return emb_standard, adaln_lora</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25Embedding
Cosmos25Embedding(embedding_dim: int, condition_dim: int, use_adaln_lora: bool = True, adaln_lora_dim: int = 256)
COSMOS 2.5 timestep conditioning embedding.
Generates sinusoidal embeddings and processes them through MLP.
Source
class Cosmos25Embedding(nn.Module):
    """
    COSMOS 2.5 timestep conditioning embedding.
    Generates sinusoidal embeddings and processes them through MLP.
    """
def __init__(
    self,
    embedding_dim: int,
    condition_dim: int,
    use_adaln_lora: bool = True,
    adaln_lora_dim: int = 256,
) -&gt; None:
    super().__init__()

    self.time_proj = Timesteps(embedding_dim, flip_sin_to_cos=True, downscale_freq_shift=0.0)
    self.t_embedder = Cosmos25TimestepEmbedding(
        embedding_dim,
        condition_dim,
        use_adaln_lora=use_adaln_lora,
        adaln_lora_dim=adaln_lora_dim,
    )
    self.norm = RMSNorm(embedding_dim, eps=1e-6)

def forward(
    self,
    hidden_states: torch.Tensor,
    timestep: torch.Tensor,
) -&gt; tuple&#91;torch.Tensor, torch.Tensor | None&#93;:
    &quot;&quot;&quot;
    Args:
        timestep: (B, T) tensor of timesteps

    Returns:
        embedded_timestep: Normalized timestep embedding (B, T, D)
        adaln_lora: AdaLN-LoRA parameters (B, T, 3D) or None
    &quot;&quot;&quot;
    # Handle 2D timestep input (B, T) like the official model
    assert timestep.ndim == 2, f&quot;Expected 2D timestep, got &#123;timestep.ndim&#125;D with shape &#123;timestep.shape&#125;&quot;
    B, T = timestep.shape

    # Flatten for Timesteps layer which expects 1D, then reshape back
    timestep_flat = timestep.flatten()  # (B*T,)
    timesteps_proj = self.time_proj(timestep_flat).type_as(hidden_states)  # (B*T, D)
    timesteps_proj = timesteps_proj.reshape(B, T, -1)  # (B, T, D)

    embedded_timestep, adaln_lora = self.t_embedder(timesteps_proj)
    embedded_timestep = self.norm(embedded_timestep)

    return embedded_timestep, adaln_lora</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25Embedding.forward
forward(hidden_states: torch.Tensor, timestep: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor | None]
Parameters

timestep: torch.Tensor
(B, T) tensor of timesteps

Returns

embedded_timestep: torch.Tensor
Normalized timestep embedding (B, T, D)
adaln_lora: torch.Tensor | None
AdaLN-LoRA parameters (B, T, 3D) or None

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    timestep: torch.Tensor,
) -> tuple[torch.Tensor, torch.Tensor | None]:
    """
    Args:
        timestep: (B, T) tensor of timesteps
Returns:
    embedded_timestep: Normalized timestep embedding (B, T, D)
    adaln_lora: AdaLN-LoRA parameters (B, T, 3D) or None
&quot;&quot;&quot;
# Handle 2D timestep input (B, T) like the official model
assert timestep.ndim == 2, f&quot;Expected 2D timestep, got &#123;timestep.ndim&#125;D with shape &#123;timestep.shape&#125;&quot;
B, T = timestep.shape

# Flatten for Timesteps layer which expects 1D, then reshape back
timestep_flat = timestep.flatten()  # (B*T,)
timesteps_proj = self.time_proj(timestep_flat).type_as(hidden_states)  # (B*T, D)
timesteps_proj = timesteps_proj.reshape(B, T, -1)  # (B, T, D)

embedded_timestep, adaln_lora = self.t_embedder(timesteps_proj)
embedded_timestep = self.norm(embedded_timestep)

return embedded_timestep, adaln_lora</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25AdaLayerNormZero
Cosmos25AdaLayerNormZero(in_features: int)
COSMOS 2.5 Adaptive Layer Normalization with zero initialization and gate.
This is a simplified version that expects pre-computed shift/scale/gate parameters.
Source
class Cosmos25AdaLayerNormZero(nn.Module):
    """
    COSMOS 2.5 Adaptive Layer Normalization with zero initialization and gate.
    This is a simplified version that expects pre-computed shift/scale/gate parameters.
    """
def __init__(
    self,
    in_features: int,
) -&gt; None:
    super().__init__()
    self.norm = nn.LayerNorm(in_features, elementwise_affine=False, eps=1e-6)

def forward(
    self,
    hidden_states: torch.Tensor,
    shift: torch.Tensor,
    scale: torch.Tensor,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: Input tensor
        shift: Shift parameter for modulation
        scale: Scale parameter for modulation

    Returns:
        normalized_hidden_states: Modulated normalized hidden states
    &quot;&quot;&quot;
    # Apply layer norm and modulation
    hidden_states = self.norm(hidden_states)
    hidden_states = hidden_states * (1 + scale) + shift
    return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25AdaLayerNormZero.forward
forward(hidden_states: torch.Tensor, shift: torch.Tensor, scale: torch.Tensor) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
Input tensor
shift: torch.Tensor
Shift parameter for modulation
scale: torch.Tensor
Scale parameter for modulation

Returns

normalized_hidden_states: torch.Tensor
Modulated normalized hidden states

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    shift: torch.Tensor,
    scale: torch.Tensor,
) -> torch.Tensor:
    """
    Args:
        hidden_states: Input tensor
        shift: Shift parameter for modulation
        scale: Scale parameter for modulation
Returns:
    normalized_hidden_states: Modulated normalized hidden states
&quot;&quot;&quot;
# Apply layer norm and modulation
hidden_states = self.norm(hidden_states)
hidden_states = hidden_states * (1 + scale) + shift
return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25SelfAttention
Cosmos25SelfAttention(dim: int, num_heads: int, qk_norm: bool = True, eps: float = 1e-06, supported_attention_backends: tuple[AttentionBackendEnum, ...] | None = None)
COSMOS 2.5 self-attention with QK normalization and RoPE.
Source
class Cosmos25SelfAttention(nn.Module):
    """
    COSMOS 2.5 self-attention with QK normalization and RoPE.
    """
def __init__(
    self,
    dim: int,
    num_heads: int,
    qk_norm: bool = True,
    eps: float = 1e-6,
    supported_attention_backends: tuple&#91;AttentionBackendEnum, ...&#93; | None = None,
) -&gt; None:
    assert dim % num_heads == 0
    super().__init__()
    self.dim = dim
    self.num_heads = num_heads
    self.head_dim = dim // num_heads

    self.to_q = ReplicatedLinear(dim, dim, bias=False)
    self.to_k = ReplicatedLinear(dim, dim, bias=False)
    self.to_v = ReplicatedLinear(dim, dim, bias=False)
    self.to_out = ReplicatedLinear(dim, dim, bias=False)

    self.norm_q = RMSNorm(self.head_dim, eps=eps) if qk_norm else nn.Identity()
    self.norm_k = RMSNorm(self.head_dim, eps=eps) if qk_norm else nn.Identity()

    # Use DistributedAttention for flexible backend support (torch SDPA / FlashAttention)
    # For single-GPU (non-distributed), use LocalAttention to avoid distributed requirements
    if supported_attention_backends is None:
        supported_attention_backends = (AttentionBackendEnum.FLASH_ATTN, AttentionBackendEnum.TORCH_SDPA)

    # Always use DistributedAttention (requires distributed environment to be initialized)
    self.attn = DistributedAttention(num_heads=num_heads,
                                     head_size=self.head_dim,
                                     causal=False,
                                     supported_attention_backends=supported_attention_backends,
                                     prefix=&quot;self_attn&quot;)

def forward(
    self,
    hidden_states: torch.Tensor,
    rope_emb: tuple&#91;torch.Tensor, torch.Tensor&#93; | None = None,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, S, D) where S = T*H*W
        rope_emb: Tuple of (cos, sin) for RoPE
    &quot;&quot;&quot;
    # Get QKV
    query, _ = self.to_q(hidden_states)
    key, _ = self.to_k(hidden_states)
    value, _ = self.to_v(hidden_states)

    # Reshape for multi-head attention: (B, S, D) -&gt; (B, S, H, D_h) -&gt; (B, H, S, D_h)
    query = query.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)
    key = key.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)
    value = value.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)

    # Apply QK normalization
    query = self.norm_q(query)
    key = self.norm_k(key)

    # Apply RoPE if provided (query/key are now in (B, H, S, D_h) format)
    if rope_emb is not None:
        cos, sin = rope_emb
        query = apply_rotary_emb(query, (cos, sin), use_real=True, use_real_unbind_dim=-2)
        key = apply_rotary_emb(key, (cos, sin), use_real=True, use_real_unbind_dim=-2)

    # Attention computation using DistributedAttention or LocalAttention
    # Both expect (B, S, H, D_h), so transpose first
    query = query.transpose(1, 2)  # (B, H, S, D_h) -&gt; (B, S, H, D_h)
    key = key.transpose(1, 2)
    value = value.transpose(1, 2)

    attn_output, _ = self.attn(query, key, value)
    # Reshape back: (B, S, H, D_h) -&gt; (B, S, H*D_h)
    attn_output = attn_output.flatten(-2, -1)

    # Output projection
    attn_output, _ = self.to_out(attn_output)
    return attn_output</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25SelfAttention.forward
forward(hidden_states: torch.Tensor, rope_emb: tuple[torch.Tensor, torch.Tensor] | None = None) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, S, D) where S = THW
rope_emb: tuple[torch.Tensor, torch.Tensor] | None
Tuple of (cos, sin) for RoPE

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    rope_emb: tuple[torch.Tensor, torch.Tensor] | None = None,
) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, S, D) where S = T*H*W
        rope_emb: Tuple of (cos, sin) for RoPE
    """
    # Get QKV
    query, _ = self.to_q(hidden_states)
    key, _ = self.to_k(hidden_states)
    value, _ = self.to_v(hidden_states)
# Reshape for multi-head attention: (B, S, D) -&gt; (B, S, H, D_h) -&gt; (B, H, S, D_h)
query = query.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)
key = key.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)
value = value.unflatten(-1, (self.num_heads, self.head_dim)).transpose(1, 2)

# Apply QK normalization
query = self.norm_q(query)
key = self.norm_k(key)

# Apply RoPE if provided (query/key are now in (B, H, S, D_h) format)
if rope_emb is not None:
    cos, sin = rope_emb
    query = apply_rotary_emb(query, (cos, sin), use_real=True, use_real_unbind_dim=-2)
    key = apply_rotary_emb(key, (cos, sin), use_real=True, use_real_unbind_dim=-2)

# Attention computation using DistributedAttention or LocalAttention
# Both expect (B, S, H, D_h), so transpose first
query = query.transpose(1, 2)  # (B, H, S, D_h) -&gt; (B, S, H, D_h)
key = key.transpose(1, 2)
value = value.transpose(1, 2)

attn_output, _ = self.attn(query, key, value)
# Reshape back: (B, S, H, D_h) -&gt; (B, S, H*D_h)
attn_output = attn_output.flatten(-2, -1)

# Output projection
attn_output, _ = self.to_out(attn_output)
return attn_output</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25CrossAttention
Cosmos25CrossAttention(dim: int, cross_attention_dim: int, num_heads: int, qk_norm: bool = True, eps: float = 1e-06, supported_attention_backends: tuple[AttentionBackendEnum, ...] | None = None)
COSMOS 2.5 cross-attention for text conditioning.
Source
class Cosmos25CrossAttention(nn.Module):
    """
    COSMOS 2.5 cross-attention for text conditioning.
    """
def __init__(
    self,
    dim: int,
    cross_attention_dim: int,
    num_heads: int,
    qk_norm: bool = True,
    eps: float = 1e-6,
    supported_attention_backends: tuple&#91;AttentionBackendEnum, ...&#93; | None = None,
) -&gt; None:
    assert dim % num_heads == 0
    super().__init__()
    self.dim = dim
    self.cross_attention_dim = cross_attention_dim
    self.num_heads = num_heads
    self.head_dim = dim // num_heads

    self.to_q = ReplicatedLinear(dim, dim, bias=False)
    self.to_k = ReplicatedLinear(cross_attention_dim, dim, bias=False)
    self.to_v = ReplicatedLinear(cross_attention_dim, dim, bias=False)
    self.to_out = ReplicatedLinear(dim, dim, bias=False)

    self.norm_q = RMSNorm(self.head_dim, eps=eps) if qk_norm else nn.Identity()
    self.norm_k = RMSNorm(self.head_dim, eps=eps) if qk_norm else nn.Identity()

    if supported_attention_backends is None:
        supported_attention_backends = (AttentionBackendEnum.FLASH_ATTN, AttentionBackendEnum.TORCH_SDPA)

    # Use LocalAttention for cross-attention since text embeddings are not sharded
    # in sequence parallelism (replicated across ranks)
    self.attn = LocalAttention(
        num_heads=num_heads,
        head_size=self.head_dim,
        causal=False,
        supported_attention_backends=supported_attention_backends,
    )

def forward(
    self,
    hidden_states: torch.Tensor,
    encoder_hidden_states: torch.Tensor,
    attention_mask: torch.Tensor | None = None,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, S, D)
        encoder_hidden_states: (B, N, D_text)
    &quot;&quot;&quot;
    # Get QKV
    query, _ = self.to_q(hidden_states)
    key, _ = self.to_k(encoder_hidden_states)
    value, _ = self.to_v(encoder_hidden_states)

    # Reshape for multi-head attention
    query = query.unflatten(-1, (self.num_heads, self.head_dim))
    key = key.unflatten(-1, (self.num_heads, self.head_dim))
    value = value.unflatten(-1, (self.num_heads, self.head_dim))

    # Apply QK normalization
    query = self.norm_q(query)
    key = self.norm_k(key)

    # LocalAttention expects (B, S, H, D_h), which is what we already have
    attn_output = self.attn(query, key, value)

    # Reshape back: (B, S, H, D_h) -&gt; (B, S, H*D_h)
    attn_output = attn_output.flatten(-2, -1)

    # Output projection
    attn_output, _ = self.to_out(attn_output)
    return attn_output</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25CrossAttention.forward
forward(hidden_states: torch.Tensor, encoder_hidden_states: torch.Tensor, attention_mask: torch.Tensor | None = None) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, S, D)
encoder_hidden_states: torch.Tensor
(B, N, D_text)

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    encoder_hidden_states: torch.Tensor,
    attention_mask: torch.Tensor | None = None,
) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, S, D)
        encoder_hidden_states: (B, N, D_text)
    """
    # Get QKV
    query, _ = self.to_q(hidden_states)
    key, _ = self.to_k(encoder_hidden_states)
    value, _ = self.to_v(encoder_hidden_states)
# Reshape for multi-head attention
query = query.unflatten(-1, (self.num_heads, self.head_dim))
key = key.unflatten(-1, (self.num_heads, self.head_dim))
value = value.unflatten(-1, (self.num_heads, self.head_dim))

# Apply QK normalization
query = self.norm_q(query)
key = self.norm_k(key)

# LocalAttention expects (B, S, H, D_h), which is what we already have
attn_output = self.attn(query, key, value)

# Reshape back: (B, S, H, D_h) -&gt; (B, S, H*D_h)
attn_output = attn_output.flatten(-2, -1)

# Output projection
attn_output, _ = self.to_out(attn_output)
return attn_output</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25TransformerBlock
Cosmos25TransformerBlock(num_attention_heads: int, attention_head_dim: int, cross_attention_dim: int, mlp_ratio: float = 4.0, adaln_lora_dim: int = 256, use_adaln_lora: bool = True, qk_norm: bool = True, supported_attention_backends: tuple[AttentionBackendEnum, ...] | None = None)
COSMOS 2.5 transformer block with self-attention, cross-attention, and MLP.
Uses AdaLN-LoRA for conditioning.
Matches the official architecture where modulation parameters are computed once per block.
Source
class Cosmos25TransformerBlock(nn.Module):
    """
    COSMOS 2.5 transformer block with self-attention, cross-attention, and MLP.
    Uses AdaLN-LoRA for conditioning.
    Matches the official architecture where modulation parameters are computed once per block.
    """
def __init__(
    self,
    num_attention_heads: int,
    attention_head_dim: int,
    cross_attention_dim: int,
    mlp_ratio: float = 4.0,
    adaln_lora_dim: int = 256,
    use_adaln_lora: bool = True,
    qk_norm: bool = True,
    supported_attention_backends: tuple&#91;AttentionBackendEnum, ...&#93; | None = None,
) -&gt; None:
    super().__init__()

    hidden_size = num_attention_heads * attention_head_dim
    self.use_adaln_lora = use_adaln_lora

    # Layer norms (no modulation logic inside)
    self.norm1 = Cosmos25AdaLayerNormZero(hidden_size)
    self.norm2 = Cosmos25AdaLayerNormZero(hidden_size)
    self.norm3 = Cosmos25AdaLayerNormZero(hidden_size)

    # Attention and MLP layers
    self.attn1 = Cosmos25SelfAttention(
        dim=hidden_size,
        num_heads=num_attention_heads,
        qk_norm=qk_norm,
        supported_attention_backends=supported_attention_backends,
    )
    self.attn2 = Cosmos25CrossAttention(
        dim=hidden_size,
        cross_attention_dim=cross_attention_dim,
        num_heads=num_attention_heads,
        qk_norm=qk_norm,
        supported_attention_backends=supported_attention_backends,
    )
    self.mlp = MLP(hidden_size, int(hidden_size * mlp_ratio), act_type=&quot;gelu&quot;, bias=False)

    # AdaLN modulation layers (compute shift/scale/gate for each sub-layer)
    # These match the official model&#x27;s adaln_modulation_* layers
    if use_adaln_lora:
        self.adaln_modulation_self_attn = nn.Sequential(
            nn.SiLU(),
            nn.Linear(hidden_size, adaln_lora_dim, bias=False),
            nn.Linear(adaln_lora_dim, 3 * hidden_size, bias=False),
        )
        self.adaln_modulation_cross_attn = nn.Sequential(
            nn.SiLU(),
            nn.Linear(hidden_size, adaln_lora_dim, bias=False),
            nn.Linear(adaln_lora_dim, 3 * hidden_size, bias=False),
        )
        self.adaln_modulation_mlp = nn.Sequential(
            nn.SiLU(),
            nn.Linear(hidden_size, adaln_lora_dim, bias=False),
            nn.Linear(adaln_lora_dim, 3 * hidden_size, bias=False),
        )
    else:
        self.adaln_modulation_self_attn = nn.Sequential(nn.SiLU(),
                                                        nn.Linear(hidden_size, 3 * hidden_size, bias=False))
        self.adaln_modulation_cross_attn = nn.Sequential(nn.SiLU(),
                                                         nn.Linear(hidden_size, 3 * hidden_size, bias=False))
        self.adaln_modulation_mlp = nn.Sequential(nn.SiLU(), nn.Linear(hidden_size, 3 * hidden_size, bias=False))

def forward(
    self,
    hidden_states: torch.Tensor,
    encoder_hidden_states: torch.Tensor,
    embedded_timestep: torch.Tensor,
    adaln_lora: torch.Tensor | None = None,
    rope_emb: tuple&#91;torch.Tensor, torch.Tensor&#93; | None = None,
    extra_pos_emb: torch.Tensor | None = None,
    attention_mask: torch.Tensor | None = None,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, T, H, W, D)
        encoder_hidden_states: (B, N, D_text)
        embedded_timestep: (B, T, D)
        adaln_lora: (B, T, 3D) AdaLN-LoRA parameters
        rope_emb: Tuple of (cos, sin) for RoPE
        extra_pos_emb: Optional learnable positional embeddings
    &quot;&quot;&quot;
    # Add extra positional embeddings if provided
    if extra_pos_emb is not None:
        hidden_states = hidden_states + extra_pos_emb

    B, T, H, W, D = hidden_states.shape

    # Step 1: Compute ALL modulation parameters once (matches official model)
    if self.use_adaln_lora and adaln_lora is not None:
        shift_self_attn, scale_self_attn, gate_self_attn = (self.adaln_modulation_self_attn(embedded_timestep) +
                                                            adaln_lora).chunk(3, dim=-1)
        shift_cross_attn, scale_cross_attn, gate_cross_attn = (self.adaln_modulation_cross_attn(embedded_timestep) +
                                                               adaln_lora).chunk(3, dim=-1)
        shift_mlp, scale_mlp, gate_mlp = (self.adaln_modulation_mlp(embedded_timestep) + adaln_lora).chunk(3,
                                                                                                           dim=-1)
    else:
        shift_self_attn, scale_self_attn, gate_self_attn = self.adaln_modulation_self_attn(embedded_timestep).chunk(
            3, dim=-1)
        shift_cross_attn, scale_cross_attn, gate_cross_attn = self.adaln_modulation_cross_attn(
            embedded_timestep).chunk(3, dim=-1)
        shift_mlp, scale_mlp, gate_mlp = self.adaln_modulation_mlp(embedded_timestep).chunk(3, dim=-1)

    # Reshape modulation parameters from (B, T, D) to (B, T, 1, 1, D) for broadcasting
    shift_self_attn = shift_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    scale_self_attn = scale_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    gate_self_attn = gate_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

    shift_cross_attn = shift_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    scale_cross_attn = scale_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    gate_cross_attn = gate_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

    shift_mlp = shift_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    scale_mlp = scale_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
    gate_mlp = gate_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

    # Step 2: Self-attention block
    norm_hidden_states = self.norm1(hidden_states, shift_self_attn, scale_self_attn)
    # Flatten for attention: (B, T, H, W, D) -&gt; (B, THW, D)
    norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

    attn_output = self.attn1(norm_hidden_states_flat, rope_emb=rope_emb)

    # Reshape back and apply residual
    attn_output = attn_output.unflatten(1, (T, H, W))  # (B, T, H, W, D)
    hidden_states = hidden_states + gate_self_attn * attn_output

    # Step 3: Cross-attention block
    norm_hidden_states = self.norm2(hidden_states, shift_cross_attn, scale_cross_attn)
    norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

    attn_output = self.attn2(
        norm_hidden_states_flat,
        encoder_hidden_states=encoder_hidden_states,
        attention_mask=attention_mask,
    )

    attn_output = attn_output.unflatten(1, (T, H, W))
    hidden_states = hidden_states + gate_cross_attn * attn_output

    # Step 4: MLP block
    norm_hidden_states = self.norm3(hidden_states, shift_mlp, scale_mlp)
    norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

    mlp_output = self.mlp(norm_hidden_states_flat)

    mlp_output = mlp_output.unflatten(1, (T, H, W))
    hidden_states = hidden_states + gate_mlp * mlp_output

    return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25TransformerBlock.forward
forward(hidden_states: torch.Tensor, encoder_hidden_states: torch.Tensor, embedded_timestep: torch.Tensor, adaln_lora: torch.Tensor | None = None, rope_emb: tuple[torch.Tensor, torch.Tensor] | None = None, extra_pos_emb: torch.Tensor | None = None, attention_mask: torch.Tensor | None = None) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, T, H, W, D)
encoder_hidden_states: torch.Tensor
(B, N, D_text)
embedded_timestep: torch.Tensor
(B, T, D)
adaln_lora: torch.Tensor | None
(B, T, 3D) AdaLN-LoRA parameters
rope_emb: tuple[torch.Tensor, torch.Tensor] | None
Tuple of (cos, sin) for RoPE
extra_pos_emb: torch.Tensor | None
Optional learnable positional embeddings

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    encoder_hidden_states: torch.Tensor,
    embedded_timestep: torch.Tensor,
    adaln_lora: torch.Tensor | None = None,
    rope_emb: tuple[torch.Tensor, torch.Tensor] | None = None,
    extra_pos_emb: torch.Tensor | None = None,
    attention_mask: torch.Tensor | None = None,
) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, T, H, W, D)
        encoder_hidden_states: (B, N, D_text)
        embedded_timestep: (B, T, D)
        adaln_lora: (B, T, 3D) AdaLN-LoRA parameters
        rope_emb: Tuple of (cos, sin) for RoPE
        extra_pos_emb: Optional learnable positional embeddings
    """
    # Add extra positional embeddings if provided
    if extra_pos_emb is not None:
        hidden_states = hidden_states + extra_pos_emb
B, T, H, W, D = hidden_states.shape

# Step 1: Compute ALL modulation parameters once (matches official model)
if self.use_adaln_lora and adaln_lora is not None:
    shift_self_attn, scale_self_attn, gate_self_attn = (self.adaln_modulation_self_attn(embedded_timestep) +
                                                        adaln_lora).chunk(3, dim=-1)
    shift_cross_attn, scale_cross_attn, gate_cross_attn = (self.adaln_modulation_cross_attn(embedded_timestep) +
                                                           adaln_lora).chunk(3, dim=-1)
    shift_mlp, scale_mlp, gate_mlp = (self.adaln_modulation_mlp(embedded_timestep) + adaln_lora).chunk(3,
                                                                                                       dim=-1)
else:
    shift_self_attn, scale_self_attn, gate_self_attn = self.adaln_modulation_self_attn(embedded_timestep).chunk(
        3, dim=-1)
    shift_cross_attn, scale_cross_attn, gate_cross_attn = self.adaln_modulation_cross_attn(
        embedded_timestep).chunk(3, dim=-1)
    shift_mlp, scale_mlp, gate_mlp = self.adaln_modulation_mlp(embedded_timestep).chunk(3, dim=-1)

# Reshape modulation parameters from (B, T, D) to (B, T, 1, 1, D) for broadcasting
shift_self_attn = shift_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
scale_self_attn = scale_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
gate_self_attn = gate_self_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

shift_cross_attn = shift_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
scale_cross_attn = scale_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
gate_cross_attn = gate_cross_attn.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

shift_mlp = shift_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
scale_mlp = scale_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)
gate_mlp = gate_mlp.unsqueeze(2).unsqueeze(2).type_as(hidden_states)

# Step 2: Self-attention block
norm_hidden_states = self.norm1(hidden_states, shift_self_attn, scale_self_attn)
# Flatten for attention: (B, T, H, W, D) -&gt; (B, THW, D)
norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

attn_output = self.attn1(norm_hidden_states_flat, rope_emb=rope_emb)

# Reshape back and apply residual
attn_output = attn_output.unflatten(1, (T, H, W))  # (B, T, H, W, D)
hidden_states = hidden_states + gate_self_attn * attn_output

# Step 3: Cross-attention block
norm_hidden_states = self.norm2(hidden_states, shift_cross_attn, scale_cross_attn)
norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

attn_output = self.attn2(
    norm_hidden_states_flat,
    encoder_hidden_states=encoder_hidden_states,
    attention_mask=attention_mask,
)

attn_output = attn_output.unflatten(1, (T, H, W))
hidden_states = hidden_states + gate_cross_attn * attn_output

# Step 4: MLP block
norm_hidden_states = self.norm3(hidden_states, shift_mlp, scale_mlp)
norm_hidden_states_flat = norm_hidden_states.flatten(1, 3)

mlp_output = self.mlp(norm_hidden_states_flat)

mlp_output = mlp_output.unflatten(1, (T, H, W))
hidden_states = hidden_states + gate_mlp * mlp_output

return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25RotaryPosEmbed
Cosmos25RotaryPosEmbed(hidden_size: int, max_size: tuple[int, int, int] = (128, 240, 240), patch_size: tuple[int, int, int] = (1, 2, 2), base_fps: int = 24, rope_scale: tuple[float, float, float] = (1.0, 1.0, 1.0), enable_fps_modulation: bool = True)
COSMOS 2.5 3D Rotary Position Embedding with NTK-aware extrapolation.
Source
class Cosmos25RotaryPosEmbed(nn.Module):
    """
    COSMOS 2.5 3D Rotary Position Embedding with NTK-aware extrapolation.
    """
def __init__(
        self,
        hidden_size: int,
        max_size: tuple&#91;int, int, int&#93; = (128, 240, 240),
        patch_size: tuple&#91;int, int, int&#93; = (1, 2, 2),
        base_fps: int = 24,
        rope_scale: tuple&#91;float, float, float&#93; = (1.0, 1.0, 1.0),
        enable_fps_modulation: bool = True,
) -&gt; None:
    super().__init__()

    self.max_size = &#91;size // patch for size, patch in zip(max_size, patch_size, strict=True)&#93;
    self.patch_size = patch_size
    self.base_fps = base_fps
    self.enable_fps_modulation = enable_fps_modulation

    # Split dimensions: 1/3 for T, 1/3 for H, 1/3 for W
    self.dim_h = hidden_size // 6 * 2
    self.dim_w = hidden_size // 6 * 2
    self.dim_t = hidden_size - self.dim_h - self.dim_w

    # NTK-aware extrapolation factors
    self.h_ntk_factor = rope_scale&#91;1&#93;**(self.dim_h / (self.dim_h - 2))
    self.w_ntk_factor = rope_scale&#91;2&#93;**(self.dim_w / (self.dim_w - 2))
    self.t_ntk_factor = rope_scale&#91;0&#93;**(self.dim_t / (self.dim_t - 2))

def forward(self, hidden_states: torch.Tensor, fps: int | None = None) -&gt; tuple&#91;torch.Tensor, torch.Tensor&#93;:
    &quot;&quot;&quot;
    Generate 3D RoPE embeddings.

    Args:
        hidden_states: (B, T, H, W, D) - patch-embedded features
        fps: Frames per second for temporal scaling

    Returns:
        cos, sin: RoPE embeddings (THW, D)
    &quot;&quot;&quot;
    batch_size, T, H, W, input_dim = hidden_states.shape
    device = hidden_states.device

    # T, H, W are already patch dimensions after patch_embed
    # No need to divide by patch_size

    # Generate frequency scales with NTK
    h_theta = 10000.0 * self.h_ntk_factor
    w_theta = 10000.0 * self.w_ntk_factor
    t_theta = 10000.0 * self.t_ntk_factor

    seq = torch.arange(max(self.max_size), device=device, dtype=torch.float32)

    # Use self.dim_h/w/t which were set during initialization
    dim_h_range = torch.arange(0, self.dim_h, 2, device=device,
                               dtype=torch.float32)&#91;:(self.dim_h // 2)&#93; / self.dim_h
    dim_w_range = torch.arange(0, self.dim_w, 2, device=device,
                               dtype=torch.float32)&#91;:(self.dim_w // 2)&#93; / self.dim_w
    dim_t_range = torch.arange(0, self.dim_t, 2, device=device,
                               dtype=torch.float32)&#91;:(self.dim_t // 2)&#93; / self.dim_t

    h_spatial_freqs = 1.0 / (h_theta**dim_h_range)
    w_spatial_freqs = 1.0 / (w_theta**dim_w_range)
    temporal_freqs = 1.0 / (t_theta**dim_t_range)

    # Generate positional embeddings
    half_emb_h = torch.outer(seq&#91;:H&#93;, h_spatial_freqs)
    half_emb_w = torch.outer(seq&#91;:W&#93;, w_spatial_freqs)

    if self.enable_fps_modulation and fps is not None:
        # Apply FPS scaling
        half_emb_t = torch.outer(seq&#91;:T&#93; / fps * self.base_fps, temporal_freqs)
    else:
        half_emb_t = torch.outer(seq&#91;:T&#93;, temporal_freqs)

    # Broadcast and concatenate embeddings
    emb_t = half_emb_t&#91;:, None, None, :&#93;.repeat(1, H, W, 1)
    emb_h = half_emb_h&#91;None, :, None, :&#93;.repeat(T, 1, W, 1)
    emb_w = half_emb_w&#91;None, None, :, :&#93;.repeat(T, H, 1, 1)

    # Concatenate &#91;t, h, w, t, h, w&#93; for sin/cos pairs
    freqs = torch.cat(&#91;emb_t, emb_h, emb_w&#93; * 2, dim=-1)
    freqs = freqs.flatten(0, 2).float()  # (THW, D)

    cos = torch.cos(freqs)  # (THW, D)
    sin = torch.sin(freqs)  # (THW, D)

    return cos, sin</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25RotaryPosEmbed.forward
forward(hidden_states: torch.Tensor, fps: int | None = None) -> tuple[torch.Tensor, torch.Tensor]
Generate 3D RoPE embeddings.
Parameters

hidden_states: torch.Tensor
(B, T, H, W, D) - patch-embedded features
fps: int | None
Frames per second for temporal scaling

Returns

tuple[torch.Tensor, torch.Tensor]
cos, sin: RoPE embeddings (THW, D)

Source
def forward(self, hidden_states: torch.Tensor, fps: int | None = None) -> tuple[torch.Tensor, torch.Tensor]:
    """
    Generate 3D RoPE embeddings.
Args:
    hidden_states: (B, T, H, W, D) - patch-embedded features
    fps: Frames per second for temporal scaling

Returns:
    cos, sin: RoPE embeddings (THW, D)
&quot;&quot;&quot;
batch_size, T, H, W, input_dim = hidden_states.shape
device = hidden_states.device

# T, H, W are already patch dimensions after patch_embed
# No need to divide by patch_size

# Generate frequency scales with NTK
h_theta = 10000.0 * self.h_ntk_factor
w_theta = 10000.0 * self.w_ntk_factor
t_theta = 10000.0 * self.t_ntk_factor

seq = torch.arange(max(self.max_size), device=device, dtype=torch.float32)

# Use self.dim_h/w/t which were set during initialization
dim_h_range = torch.arange(0, self.dim_h, 2, device=device,
                           dtype=torch.float32)&#91;:(self.dim_h // 2)&#93; / self.dim_h
dim_w_range = torch.arange(0, self.dim_w, 2, device=device,
                           dtype=torch.float32)&#91;:(self.dim_w // 2)&#93; / self.dim_w
dim_t_range = torch.arange(0, self.dim_t, 2, device=device,
                           dtype=torch.float32)&#91;:(self.dim_t // 2)&#93; / self.dim_t

h_spatial_freqs = 1.0 / (h_theta**dim_h_range)
w_spatial_freqs = 1.0 / (w_theta**dim_w_range)
temporal_freqs = 1.0 / (t_theta**dim_t_range)

# Generate positional embeddings
half_emb_h = torch.outer(seq&#91;:H&#93;, h_spatial_freqs)
half_emb_w = torch.outer(seq&#91;:W&#93;, w_spatial_freqs)

if self.enable_fps_modulation and fps is not None:
    # Apply FPS scaling
    half_emb_t = torch.outer(seq&#91;:T&#93; / fps * self.base_fps, temporal_freqs)
else:
    half_emb_t = torch.outer(seq&#91;:T&#93;, temporal_freqs)

# Broadcast and concatenate embeddings
emb_t = half_emb_t&#91;:, None, None, :&#93;.repeat(1, H, W, 1)
emb_h = half_emb_h&#91;None, :, None, :&#93;.repeat(T, 1, W, 1)
emb_w = half_emb_w&#91;None, None, :, :&#93;.repeat(T, H, 1, 1)

# Concatenate &#91;t, h, w, t, h, w&#93; for sin/cos pairs
freqs = torch.cat(&#91;emb_t, emb_h, emb_w&#93; * 2, dim=-1)
freqs = freqs.flatten(0, 2).float()  # (THW, D)

cos = torch.cos(freqs)  # (THW, D)
sin = torch.sin(freqs)  # (THW, D)

return cos, sin</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25LearnablePositionalEmbed
Cosmos25LearnablePositionalEmbed(hidden_size: int, max_size: tuple[int, int, int], patch_size: tuple[int, int, int], eps: float = 1e-06)
COSMOS 2.5 learnable absolute positional embeddings (optional).
Source
class Cosmos25LearnablePositionalEmbed(nn.Module):
    """
    COSMOS 2.5 learnable absolute positional embeddings (optional).
    """
def __init__(
    self,
    hidden_size: int,
    max_size: tuple&#91;int, int, int&#93;,
    patch_size: tuple&#91;int, int, int&#93;,
    eps: float = 1e-6,
) -&gt; None:
    super().__init__()

    self.max_size = &#91;size // patch for size, patch in zip(max_size, patch_size, strict=True)&#93;
    self.patch_size = patch_size
    self.eps = eps

    self.pos_emb_t = nn.Parameter(torch.zeros(self.max_size&#91;0&#93;, hidden_size))
    self.pos_emb_h = nn.Parameter(torch.zeros(self.max_size&#91;1&#93;, hidden_size))
    self.pos_emb_w = nn.Parameter(torch.zeros(self.max_size&#91;2&#93;, hidden_size))

def forward(self, hidden_states: torch.Tensor) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, T, H, W, D)
    Returns:
        pos_emb: (B, T, H, W, D)
    &quot;&quot;&quot;
    B, T, H, W, D = hidden_states.shape

    emb_t = self.pos_emb_t&#91;:T&#93;&#91;None, :, None, None, :&#93;.repeat(B, 1, H, W, 1)
    emb_h = self.pos_emb_h&#91;:H&#93;&#91;None, None, :, None, :&#93;.repeat(B, T, 1, W, 1)
    emb_w = self.pos_emb_w&#91;:W&#93;&#91;None, None, None, :, :&#93;.repeat(B, T, H, 1, 1)

    emb = emb_t + emb_h + emb_w

    # Normalize
    norm = torch.linalg.vector_norm(emb, dim=-1, keepdim=True, dtype=torch.float32)
    norm = torch.add(self.eps, norm, alpha=np.sqrt(norm.numel() / emb.numel()))
    return (emb / norm).type_as(hidden_states)</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25LearnablePositionalEmbed.forward
forward(hidden_states: torch.Tensor) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, T, H, W, D)

Returns:
    pos_emb: (B, T, H, W, D)
Source
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, T, H, W, D)
    Returns:
        pos_emb: (B, T, H, W, D)
    """
    B, T, H, W, D = hidden_states.shape
emb_t = self.pos_emb_t&#91;:T&#93;&#91;None, :, None, None, :&#93;.repeat(B, 1, H, W, 1)
emb_h = self.pos_emb_h&#91;:H&#93;&#91;None, None, :, None, :&#93;.repeat(B, T, 1, W, 1)
emb_w = self.pos_emb_w&#91;:W&#93;&#91;None, None, None, :, :&#93;.repeat(B, T, H, 1, 1)

emb = emb_t + emb_h + emb_w

# Normalize
norm = torch.linalg.vector_norm(emb, dim=-1, keepdim=True, dtype=torch.float32)
norm = torch.add(self.eps, norm, alpha=np.sqrt(norm.numel() / emb.numel()))
return (emb / norm).type_as(hidden_states)</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25FinalLayer
Cosmos25FinalLayer(hidden_size: int, out_channels: int, patch_size: tuple[int, int, int], adaln_lora_dim: int = 256, use_adaln_lora: bool = True)
COSMOS 2.5 final layer with AdaLN modulation and unpatchification.
Source
class Cosmos25FinalLayer(nn.Module):
    """
    COSMOS 2.5 final layer with AdaLN modulation and unpatchification.
    """
def __init__(
    self,
    hidden_size: int,
    out_channels: int,
    patch_size: tuple&#91;int, int, int&#93;,
    adaln_lora_dim: int = 256,
    use_adaln_lora: bool = True,
) -&gt; None:
    super().__init__()
    self.hidden_size = hidden_size
    self.use_adaln_lora = use_adaln_lora

    self.norm = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
    self.activation = nn.SiLU()

    if use_adaln_lora:
        self.linear_1 = nn.Linear(hidden_size, adaln_lora_dim, bias=False)
        self.linear_2 = nn.Linear(adaln_lora_dim, 2 * hidden_size, bias=False)
    else:
        self.linear_1 = nn.Identity()
        self.linear_2 = nn.Linear(hidden_size, 2 * hidden_size, bias=False)

    # Output projection
    output_dim = out_channels * patch_size&#91;0&#93; * patch_size&#91;1&#93; * patch_size&#91;2&#93;
    self.proj_out = nn.Linear(hidden_size, output_dim, bias=False)

def forward(
    self,
    hidden_states: torch.Tensor,
    embedded_timestep: torch.Tensor,
    adaln_lora: torch.Tensor | None = None,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, T, H, W, D)
        embedded_timestep: (B, T, D) or (B, D)
        adaln_lora: (B, T, 3D) or None
    &quot;&quot;&quot;
    # Generate modulation parameters
    embedded_timestep = self.activation(embedded_timestep)
    embedded_timestep = self.linear_1(embedded_timestep)
    embedded_timestep = self.linear_2(embedded_timestep)

    if self.use_adaln_lora and adaln_lora is not None:
        # Use first 2*hidden_size elements for shift/scale
        embedded_timestep = embedded_timestep + adaln_lora&#91;..., :2 * self.hidden_size&#93;

    shift, scale = embedded_timestep.chunk(2, dim=-1)

    # Apply normalization and modulation
    hidden_states = self.norm(hidden_states)

    # Reshape for broadcasting if needed
    if embedded_timestep.ndim == 2:
        shift, scale = (x.unsqueeze(1) for x in (shift, scale))
    elif embedded_timestep.ndim == 3 and hidden_states.ndim == 5:
        shift, scale = (x.unsqueeze(2).unsqueeze(2) for x in (shift, scale))

    hidden_states = hidden_states * (1 + scale) + shift

    # Project to output
    hidden_states = self.proj_out(hidden_states)

    return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25FinalLayer.forward
forward(hidden_states: torch.Tensor, embedded_timestep: torch.Tensor, adaln_lora: torch.Tensor | None = None) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, T, H, W, D)
embedded_timestep: torch.Tensor
(B, T, D) or (B, D)
adaln_lora: torch.Tensor | None
(B, T, 3D) or None

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    embedded_timestep: torch.Tensor,
    adaln_lora: torch.Tensor | None = None,
) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, T, H, W, D)
        embedded_timestep: (B, T, D) or (B, D)
        adaln_lora: (B, T, 3D) or None
    """
    # Generate modulation parameters
    embedded_timestep = self.activation(embedded_timestep)
    embedded_timestep = self.linear_1(embedded_timestep)
    embedded_timestep = self.linear_2(embedded_timestep)
if self.use_adaln_lora and adaln_lora is not None:
    # Use first 2*hidden_size elements for shift/scale
    embedded_timestep = embedded_timestep + adaln_lora&#91;..., :2 * self.hidden_size&#93;

shift, scale = embedded_timestep.chunk(2, dim=-1)

# Apply normalization and modulation
hidden_states = self.norm(hidden_states)

# Reshape for broadcasting if needed
if embedded_timestep.ndim == 2:
    shift, scale = (x.unsqueeze(1) for x in (shift, scale))
elif embedded_timestep.ndim == 3 and hidden_states.ndim == 5:
    shift, scale = (x.unsqueeze(2).unsqueeze(2) for x in (shift, scale))

hidden_states = hidden_states * (1 + scale) + shift

# Project to output
hidden_states = self.proj_out(hidden_states)

return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25Transformer3DModel
Cosmos25Transformer3DModel(config: Cosmos25VideoConfig, hf_config: dict[str, Any])
COSMOS 2.5 DiT - MiniTrainDIT architecture adapted for FastVideo.
Key features:
- AdaLN-LoRA conditioning
- 3D RoPE with NTK-aware extrapolation
- Optional learnable positional embeddings
- QK normalization
- Cross-attention projection (optional)
Source
class Cosmos25Transformer3DModel(BaseDiT):
    """
    COSMOS 2.5 DiT - MiniTrainDIT architecture adapted for FastVideo.
Key features:
- AdaLN-LoRA conditioning
- 3D RoPE with NTK-aware extrapolation
- Optional learnable positional embeddings
- QK normalization
- Cross-attention projection (optional)
&quot;&quot;&quot;

_fsdp_shard_conditions = Cosmos25VideoConfig()._fsdp_shard_conditions
_compile_conditions = Cosmos25VideoConfig()._compile_conditions
param_names_mapping = Cosmos25VideoConfig().param_names_mapping
lora_param_names_mapping = Cosmos25VideoConfig().lora_param_names_mapping

def __init__(self, config: Cosmos25VideoConfig, hf_config: dict&#91;str, Any&#93;) -&gt; None:
    super().__init__(config=config, hf_config=hf_config)

    inner_dim = config.num_attention_heads * config.attention_head_dim
    self.hidden_size = inner_dim
    self.num_attention_heads = config.num_attention_heads
    self.in_channels = config.in_channels
    self.out_channels = config.out_channels
    self.num_channels_latents = config.num_channels_latents
    self.patch_size = config.patch_size
    self.max_size = config.max_size
    self.rope_scale = config.rope_scale
    self.concat_padding_mask = config.concat_padding_mask
    self.use_adaln_lora = getattr(config, &quot;use_adaln_lora&quot;, True)
    self.adaln_lora_dim = getattr(config, &quot;adaln_lora_dim&quot;, 256)
    self.extra_pos_embed_type = getattr(config, &quot;extra_pos_embed_type&quot;, None)
    self.use_crossattn_projection = getattr(config, &quot;use_crossattn_projection&quot;, False)

    # 1. Patch Embedding
    # Account for: VAE channels + condition_mask (1) + padding_mask (1 if concat_padding_mask)
    patch_embed_in_channels = config.in_channels  # Base VAE channels (16)
    patch_embed_in_channels += 1  # Always add 1 for condition_mask
    if config.concat_padding_mask:
        patch_embed_in_channels += 1  # Add 1 for padding_mask
    # Total: 16 + 1 + 1 = 18 (with concat_padding_mask=True)

    self.patch_embed = Cosmos25PatchEmbed(patch_embed_in_channels, inner_dim, config.patch_size)

    # 2. Positional Embeddings
    self.rope = Cosmos25RotaryPosEmbed(
        hidden_size=config.attention_head_dim,
        max_size=config.max_size,
        patch_size=config.patch_size,
        rope_scale=config.rope_scale,
        enable_fps_modulation=getattr(config, &quot;rope_enable_fps_modulation&quot;, True),
    )

    self.learnable_pos_embed = None
    if self.extra_pos_embed_type == &quot;learnable&quot;:
        self.learnable_pos_embed = Cosmos25LearnablePositionalEmbed(
            hidden_size=inner_dim,
            max_size=config.max_size,
            patch_size=config.patch_size,
        )

    # 3. Time Embedding
    self.time_embed = Cosmos25Embedding(
        inner_dim,
        inner_dim,
        use_adaln_lora=self.use_adaln_lora,
        adaln_lora_dim=self.adaln_lora_dim,
    )

    # 4. Cross-attention projection (optional)
    if self.use_crossattn_projection:
        crossattn_proj_in_channels = getattr(config, &quot;crossattn_proj_in_channels&quot;, config.text_embed_dim)
        self.crossattn_proj = nn.Sequential(
            nn.Linear(crossattn_proj_in_channels, config.text_embed_dim, bias=True),
            nn.GELU(),
        )

    # 5. Transformer Blocks
    self.transformer_blocks = nn.ModuleList(&#91;
        Cosmos25TransformerBlock(
            num_attention_heads=config.num_attention_heads,
            attention_head_dim=config.attention_head_dim,
            cross_attention_dim=config.text_embed_dim,
            mlp_ratio=config.mlp_ratio,
            adaln_lora_dim=self.adaln_lora_dim,
            use_adaln_lora=self.use_adaln_lora,
            qk_norm=(config.qk_norm == &quot;rms_norm&quot;),
            supported_attention_backends=config._supported_attention_backends,
        ) for i in range(config.num_layers)
    &#93;)

    # 6. Final Layer
    self.final_layer = Cosmos25FinalLayer(
        hidden_size=inner_dim,
        out_channels=config.out_channels,
        patch_size=config.patch_size,
        adaln_lora_dim=self.adaln_lora_dim,
        use_adaln_lora=self.use_adaln_lora,
    )

    self.gradient_checkpointing = False
    self.__post_init__()

def forward(
    self,
    hidden_states: torch.Tensor,
    timestep: torch.Tensor,
    encoder_hidden_states: torch.Tensor | list&#91;torch.Tensor&#93;,
    attention_mask: torch.Tensor | None = None,
    fps: int | None = None,
    condition_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    **kwargs,
) -&gt; torch.Tensor:
    &quot;&quot;&quot;
    Args:
        hidden_states: (B, C, T, H, W) latent video
        timestep: (B,) or (B, T) diffusion timesteps
        encoder_hidden_states: (B, N, D_text) text embeddings
        attention_mask: Optional attention mask
        fps: Frames per second
        condition_mask: (B, 1, T, H, W) conditioning mask
        padding_mask: (B, 1, H, W) padding mask
    &quot;&quot;&quot;
    if not isinstance(encoder_hidden_states, torch.Tensor):
        encoder_hidden_states = encoder_hidden_states&#91;0&#93;

    batch_size, num_channels, num_frames, height, width = hidden_states.shape

    # 1. Concatenate condition mask if provided
    if condition_mask is not None:
        hidden_states = torch.cat(&#91;hidden_states, condition_mask&#93;, dim=1)

    # 2. Concatenate padding mask if needed
    if self.concat_padding_mask and padding_mask is not None:
        padding_mask = transforms.functional.resize(
            padding_mask,
            list(hidden_states.shape&#91;-2:&#93;),
            interpolation=transforms.InterpolationMode.NEAREST,
        )
        hidden_states = torch.cat(
            &#91;hidden_states, padding_mask.unsqueeze(2).repeat(1, 1, num_frames, 1, 1)&#93;,
            dim=1,
        )

    # 3. Patchify input
    p_t, p_h, p_w = self.patch_size
    post_patch_num_frames = num_frames // p_t
    post_patch_height = height // p_h
    post_patch_width = width // p_w

    hidden_states = self.patch_embed(hidden_states)  # (B, T&#x27;, H&#x27;, W&#x27;, D)

    # 4. Generate RoPE embeddings (after patchify, using patch dimensions)
    rope_emb = self.rope(hidden_states, fps=fps)

    # 5. Generate learnable positional embeddings (if used)
    extra_pos_emb = None
    if self.learnable_pos_embed is not None:
        extra_pos_emb = self.learnable_pos_embed(hidden_states)

    # 6. Timestep embeddings
    # Official model expects timestep in (B, T) format, so ensure it has 2D shape
    if timestep.ndim == 1:
        # Scalar timestep per sample: (B,) -&gt; (B, 1)
        timestep = timestep.unsqueeze(1)
    elif timestep.ndim == 2:
        # Already in (B, T) format
        pass
    else:
        raise ValueError(f&quot;Unsupported timestep shape: &#123;timestep.shape&#125;&quot;)

    # Now timestep is always (B, T), pass directly to time_embed
    embedded_timestep, adaln_lora = self.time_embed(hidden_states, timestep)

    # 7. Apply cross-attention projection (if used)
    if self.use_crossattn_projection:
        encoder_hidden_states = self.crossattn_proj(encoder_hidden_states.to(self.crossattn_proj&#91;0&#93;.weight.dtype))

    # Prepare attention mask
    if attention_mask is not None:
        attention_mask = attention_mask.unsqueeze(1).unsqueeze(1)  # (B, 1, 1, N)

    # 8. Transformer blocks
    if torch.is_grad_enabled() and self.gradient_checkpointing:
        for block in self.transformer_blocks:
            hidden_states = self._gradient_checkpointing_func(
                block,
                hidden_states,
                encoder_hidden_states,
                embedded_timestep,
                adaln_lora,
                rope_emb,
                extra_pos_emb,
                attention_mask,
            )
    else:
        for i, block in enumerate(self.transformer_blocks):
            hidden_states = block(
                hidden_states=hidden_states,
                encoder_hidden_states=encoder_hidden_states,
                embedded_timestep=embedded_timestep,
                adaln_lora=adaln_lora,
                rope_emb=rope_emb,
                extra_pos_emb=extra_pos_emb,
                attention_mask=attention_mask,
            )

    # 9. Final layer - output norm &amp; projection
    hidden_states = self.final_layer(hidden_states, embedded_timestep, adaln_lora)

    # 10. Unpatchify: (B, T&#x27;, H&#x27;, W&#x27;, P) -&gt; (B, C, T, H, W)
    # After unflatten: (B, T&#x27;, H&#x27;, W&#x27;, p_t, p_h, p_w, C) with dims &#91;0,1,2,3,4,5,6,7&#93;
    hidden_states = hidden_states.unflatten(-1, (p_t, p_h, p_w, self.out_channels))
    # Permute to: (B, C, T&#x27;, p_t, H&#x27;, p_h, W&#x27;, p_w)
    hidden_states = hidden_states.permute(0, 7, 1, 4, 2, 5, 3, 6)
    # Flatten pairs to get (B, C, T, H, W)
    hidden_states = hidden_states.flatten(2, 3).flatten(3, 4).flatten(4, 5)

    return hidden_states</code></pre></details>
fastvideo.models.dits.cosmos2_5.Cosmos25Transformer3DModel.forward
forward(hidden_states: torch.Tensor, timestep: torch.Tensor, encoder_hidden_states: torch.Tensor | list[torch.Tensor], attention_mask: torch.Tensor | None = None, fps: int | None = None, condition_mask: torch.Tensor | None = None, padding_mask: torch.Tensor | None = None, **kwargs) -> torch.Tensor
Parameters

hidden_states: torch.Tensor
(B, C, T, H, W) latent video
timestep: torch.Tensor
(B,) or (B, T) diffusion timesteps
encoder_hidden_states: torch.Tensor | list[torch.Tensor]
(B, N, D_text) text embeddings
attention_mask: torch.Tensor | None
Optional attention mask
fps: int | None
Frames per second
condition_mask: torch.Tensor | None
(B, 1, T, H, W) conditioning mask
padding_mask: torch.Tensor | None
(B, 1, H, W) padding mask

Source
def forward(
    self,
    hidden_states: torch.Tensor,
    timestep: torch.Tensor,
    encoder_hidden_states: torch.Tensor | list[torch.Tensor],
    attention_mask: torch.Tensor | None = None,
    fps: int | None = None,
    condition_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    **kwargs,
) -> torch.Tensor:
    """
    Args:
        hidden_states: (B, C, T, H, W) latent video
        timestep: (B,) or (B, T) diffusion timesteps
        encoder_hidden_states: (B, N, D_text) text embeddings
        attention_mask: Optional attention mask
        fps: Frames per second
        condition_mask: (B, 1, T, H, W) conditioning mask
        padding_mask: (B, 1, H, W) padding mask
    """
    if not isinstance(encoder_hidden_states, torch.Tensor):
        encoder_hidden_states = encoder_hidden_states[0]
batch_size, num_channels, num_frames, height, width = hidden_states.shape

# 1. Concatenate condition mask if provided
if condition_mask is not None:
    hidden_states = torch.cat(&#91;hidden_states, condition_mask&#93;, dim=1)

# 2. Concatenate padding mask if needed
if self.concat_padding_mask and padding_mask is not None:
    padding_mask = transforms.functional.resize(
        padding_mask,
        list(hidden_states.shape&#91;-2:&#93;),
        interpolation=transforms.InterpolationMode.NEAREST,
    )
    hidden_states = torch.cat(
        &#91;hidden_states, padding_mask.unsqueeze(2).repeat(1, 1, num_frames, 1, 1)&#93;,
        dim=1,
    )

# 3. Patchify input
p_t, p_h, p_w = self.patch_size
post_patch_num_frames = num_frames // p_t
post_patch_height = height // p_h
post_patch_width = width // p_w

hidden_states = self.patch_embed(hidden_states)  # (B, T&#x27;, H&#x27;, W&#x27;, D)

# 4. Generate RoPE embeddings (after patchify, using patch dimensions)
rope_emb = self.rope(hidden_states, fps=fps)

# 5. Generate learnable positional embeddings (if used)
extra_pos_emb = None
if self.learnable_pos_embed is not None:
    extra_pos_emb = self.learnable_pos_embed(hidden_states)

# 6. Timestep embeddings
# Official model expects timestep in (B, T) format, so ensure it has 2D shape
if timestep.ndim == 1:
    # Scalar timestep per sample: (B,) -&gt; (B, 1)
    timestep = timestep.unsqueeze(1)
elif timestep.ndim == 2:
    # Already in (B, T) format
    pass
else:
    raise ValueError(f&quot;Unsupported timestep shape: &#123;timestep.shape&#125;&quot;)

# Now timestep is always (B, T), pass directly to time_embed
embedded_timestep, adaln_lora = self.time_embed(hidden_states, timestep)

# 7. Apply cross-attention projection (if used)
if self.use_crossattn_projection:
    encoder_hidden_states = self.crossattn_proj(encoder_hidden_states.to(self.crossattn_proj&#91;0&#93;.weight.dtype))

# Prepare attention mask
if attention_mask is not None:
    attention_mask = attention_mask.unsqueeze(1).unsqueeze(1)  # (B, 1, 1, N)

# 8. Transformer blocks
if torch.is_grad_enabled() and self.gradient_checkpointing:
    for block in self.transformer_blocks:
        hidden_states = self._gradient_checkpointing_func(
            block,
            hidden_states,
            encoder_hidden_states,
            embedded_timestep,
            adaln_lora,
            rope_emb,
            extra_pos_emb,
            attention_mask,
        )
else:
    for i, block in enumerate(self.transformer_blocks):
        hidden_states = block(
            hidden_states=hidden_states,
            encoder_hidden_states=encoder_hidden_states,
            embedded_timestep=embedded_timestep,
            adaln_lora=adaln_lora,
            rope_emb=rope_emb,
            extra_pos_emb=extra_pos_emb,
            attention_mask=attention_mask,
        )

# 9. Final layer - output norm &amp; projection
hidden_states = self.final_layer(hidden_states, embedded_timestep, adaln_lora)

# 10. Unpatchify: (B, T&#x27;, H&#x27;, W&#x27;, P) -&gt; (B, C, T, H, W)
# After unflatten: (B, T&#x27;, H&#x27;, W&#x27;, p_t, p_h, p_w, C) with dims &#91;0,1,2,3,4,5,6,7&#93;
hidden_states = hidden_states.unflatten(-1, (p_t, p_h, p_w, self.out_channels))
# Permute to: (B, C, T&#x27;, p_t, H&#x27;, p_h, W&#x27;, p_w)
hidden_states = hidden_states.permute(0, 7, 1, 4, 2, 5, 3, 6)
# Flatten pairs to get (B, C, T, H, W)
hidden_states = hidden_states.flatten(2, 3).flatten(3, 4).flatten(4, 5)

return hidden_states</code></pre></details>