( self, in_channels: int, out_channels: int, temb_channels: int, prev_

(self, model_dims: int, num_heads: int, activation: str)

(self, bpe_ranks, vocab)

(self, model: str = _DEFAULT_MODEL, float16: bool = False)

(self, dims: int, norm_groups: int = 32)

( self, in_channels: int, out_channels: int, num_layers: int = 1, resn

( self, in_channels: int, out_channels: int, block_out_channels: List[int] = [

(self, config: AutoencoderConfig)

(self, config: DiffusionConfig)

(self, config, model_name)

(self, config, bidirectional: bool)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

( self, config, in_channels: int, out_channels: int, kernel_size: int,

( self, config, in_channels: int, out_channels: int, kernel_size: int,

(self, config, dimension)

(self, config, dim: int, dilations: List[int])

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, t5_name, input_dim, output_dim)

(self, dim, n_heads)

(self, config)

(self, config)

(self, in_channels, out_channels, kernel_size, stride, padding)

(self, num_latent_dims, image_shape, max_num_filters)

(self, num_latent_dims, image_shape, max_num_filters)

(self, tokenizer_path: str)

(self, dim: int = 1280, num_heads: int = 16, mlp_ratio: int = 4)

(self, dim: int, num_heads: int)

( self, image_size: int = 224, patch_size: int = 14, dim: int = 1280,

( self, dim: int, num_heads: int, eps: float = 1e-6, )

( self, dim: int, num_heads: int, eps: float = 1e-6, )

(self, dim: int, num_heads: int, eps: float = 1e-6)

( self, dim: int, ffn_dim: int, num_heads: int, cross_attn_norm: bool

(self, num_buckets, num_heads, bidirectional=True, max_dist=128)

(self, dim, dim_attn, num_heads)

(self, dim, dim_ffn)

(self, dim, dim_attn, dim_ffn, num_heads, num_buckets, shared_pos=True)

( self, dim: int = 96, z_dim: int = 16, dim_mult: Optional[List[int]] = None,

( self, dim: int = 96, z_dim: int = 16, dim_mult: Optional[List[int]] = None,

( self, name: str = "t2v-1.3B", dtype: mx.Dtype = mx.bfloat16, checkpoint: Opt

( self, num_train_timesteps: int = 1000, solver_order: int = 2, prediction_typ

(self, num_train_timesteps=1000)

( self, in_channels, out_channels, kernel_size, stride=1, padding=0, bias=True )

(self, in_dim, out_dim)

(self, dim)

( self, model_type: str = "t2v", patch_size: Tuple[int, int, int] = (1, 2, 2),

(self, config, model_name)

(self, config, bidirectional: bool)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, config)

(self, in_features, out_features, bias=True)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self, key: str, item: Any)

( self, x: Tensor, weight: Tensor, bias: Optional[Tensor] )

(self)

(inputs, audio_features, tokens, sum_logprobs)

(self)

(self)

(self, logits: mx.array, tokens: mx.array)

(self, logits: mx.array, tokens: mx.array)

(self, logits: mx.array, tokens: mx.array)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(self)

(seg, expected)

(words, expected_words)

(self)

(self)