Hugging Face's logo

Transformers documentation

Cosmos3 Edge

You are viewing main version, which requires installation from source. If you'd like regular pip install, checkout the latest stable version (v5.14.0).
Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

This model was contributed to Hugging Face Transformers on 2026-07-16.

Cosmos3 Edge

Cosmos3 Edge is NVIDIA’s multimodal reasoning model from the Cosmos3 family. Transformers integrates the Reasoner tower only; the checkpoint’s diffusion Generator, VAE, scheduler, and other generation components remain Diffusers components.

The reasoner uses a dense, Llama-compatible language tower with 28 decoder blocks, each containing attention and an MLP. Its SigLIP2 vision encoder accepts packed variable-resolution patches, uses sequence boundaries to keep images and video frames independent during vision attention, groups patches spatially in 2×2 blocks, and projects them into the language model. Image and video inputs use multimodal rotary position IDs; video prompts are expanded into one timestamped vision span per sampled frame.

Usage

from transformers import AutoModelForImageTextToText, AutoProcessor

model_id = "nvidia/Cosmos3-Edge"
model = AutoModelForImageTextToText.from_pretrained(model_id, device_map="auto")
processor = AutoProcessor.from_pretrained(model_id)

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
            },
            {"type": "text", "text": "Describe this image."},
        ],
    }
]

inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt",
).to(model.device)

generated_ids = model.generate(**inputs, max_new_tokens=128)
generated_ids = [output_ids[len(input_ids) :] for input_ids, output_ids in zip(inputs.input_ids, generated_ids)]
print(processor.batch_decode(generated_ids, skip_special_tokens=True))

Cosmos3EdgeConfig

class transformers.Cosmos3EdgeConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonetext_config: transformers.models.cosmos3_edge.configuration_cosmos3_edge.Cosmos3EdgeTextConfig | dict | None = Nonevision_config: transformers.models.cosmos3_edge.configuration_cosmos3_edge.Cosmos3EdgeVisionConfig | dict | None = Noneprojector_hidden_size: int = 11520image_token_id: int = 19video_token_id: int = 18vision_start_token_id: int = 20vision_end_token_id: int = 21tie_word_embeddings: bool = False )

Parameters

  • text_config (Union[~models.cosmos3_edge.configuration_cosmos3_edge.Cosmos3EdgeTextConfig, dict], optional) — The config object or dictionary of the text backbone.
  • vision_config (Union[~models.cosmos3_edge.configuration_cosmos3_edge.Cosmos3EdgeVisionConfig, dict], optional) — The config object or dictionary of the vision backbone.
  • projector_hidden_size (int, optional, defaults to 11520) — Intermediate hidden size of the vision-to-language projector MLP.
  • image_token_id (int, optional, defaults to 19) — The image token index used as a placeholder for input images.
  • video_token_id (int, optional, defaults to 18) — The video token index used as a placeholder for input videos.
  • vision_start_token_id (int, optional, defaults to 20) — Token ID that marks the start of a visual segment in the multimodal input sequence.
  • vision_end_token_id (int, optional, defaults to 21) — Token ID that marks the end of a visual segment in the multimodal input sequence.
  • tie_word_embeddings (bool, optional, defaults to False) — Whether to tie weight embeddings according to model’s tied_weights_keys mapping.

This is the configuration class to store the configuration of a Cosmos3EdgeModel. It is used to instantiate a Cosmos3 Edge model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the nvidia/Cosmos3-Edge-Reasoner

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Example:

>>> from transformers import Cosmos3EdgeConfig, Cosmos3EdgeForConditionalGeneration

>>> configuration = Cosmos3EdgeConfig()
>>> model = Cosmos3EdgeForConditionalGeneration(configuration)

Cosmos3EdgeTextConfig

class transformers.Cosmos3EdgeTextConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonevocab_size: int = 131072hidden_size: int = 2048intermediate_size: int = 9216num_hidden_layers: int = 28num_attention_heads: int = 16num_key_value_heads: int | None = 8hidden_act: str = 'relu2'max_position_embeddings: int = 131072initializer_range: float = 0.02rms_norm_eps: float = 1e-05use_cache: bool = Truepad_token_id: int | None = 0bos_token_id: int | None = 1eos_token_id: int | list[int] | None = 11pretraining_tp: int | None = 1tie_word_embeddings: bool = Falserope_parameters: dict | None = Noneattention_bias: bool = Falseattention_dropout: float | int = 0.0mlp_bias: bool = Falsehead_dim: int = 128 )

Parameters

  • vocab_size (int, optional, defaults to 131072) — Vocabulary size of the model. Defines the number of different tokens that can be represented by the input_ids.
  • hidden_size (int, optional, defaults to 2048) — Dimension of the hidden representations.
  • intermediate_size (int, optional, defaults to 9216) — Dimension of the MLP representations.
  • num_hidden_layers (int, optional, defaults to 28) — Number of hidden layers in the Transformer decoder.
  • num_attention_heads (int, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer decoder.
  • num_key_value_heads (int, optional, defaults to 8) — This is the number of key_value heads that should be used to implement Grouped Query Attention. If num_key_value_heads=num_attention_heads, the model will use Multi Head Attention (MHA), if num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out this paper. If it is not specified, will default to num_attention_heads.
  • hidden_act (str, optional, defaults to relu2) — The non-linear activation function (function or string) in the decoder. For example, "gelu", "relu", "silu", etc.
  • max_position_embeddings (int, optional, defaults to 131072) — The maximum sequence length that this model might ever be used with.
  • initializer_range (float, optional, defaults to 0.02) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • rms_norm_eps (float, optional, defaults to 1e-05) — The epsilon used by the rms normalization layers.
  • use_cache (bool, optional, defaults to True) — Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if config.is_decoder=True or when the model is a decoder-only generative model.
  • pad_token_id (int, optional, defaults to 0) — Token id used for padding in the vocabulary.
  • bos_token_id (int, optional, defaults to 1) — Token id used for beginning-of-stream in the vocabulary.
  • eos_token_id (Union[int, list[int]], optional, defaults to 11) — Token id used for end-of-stream in the vocabulary.
  • pretraining_tp (int, optional, defaults to 1) — Experimental feature. Tensor parallelism rank used during pretraining. Please refer to this document to understand more about it. This value is necessary to ensure exact reproducibility of the pretraining results. Please refer to this issue.
  • tie_word_embeddings (bool, optional, defaults to False) — Whether to tie weight embeddings according to model’s tied_weights_keys mapping.
  • rope_parameters (dict, optional) — Dictionary containing the configuration parameters for the RoPE embeddings. The dictionary should contain a value for rope_theta and optionally parameters used for scaling in case you want to use RoPE with longer max_position_embeddings.
  • attention_bias (bool, optional, defaults to False) — Whether to use a bias in the query, key, value and output projection layers during self-attention.
  • attention_dropout (Union[float, int], optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
  • mlp_bias (bool, optional, defaults to False) — Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP layers.
  • head_dim (int, optional, defaults to 128) — The attention head dimension. If None, it will default to hidden_size // num_attention_heads

This is the configuration class to store the configuration of a Cosmos3EdgeModel. It is used to instantiate a Cosmos3 Edge model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the nvidia/Cosmos3-Edge-Reasoner

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

>>> from transformers import Cosmos3EdgeTextModel, Cosmos3EdgeTextConfig

>>> # Initializing a Cosmos3EdgeText cosmos3_edge_text-7b style configuration
>>> configuration = Cosmos3EdgeTextConfig()

>>> # Initializing a model from the cosmos3_edge_text-7b style configuration
>>> model = Cosmos3EdgeTextModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

Cosmos3EdgeVisionConfig

class transformers.Cosmos3EdgeVisionConfig

< >

( transformers_version: str | None = Nonearchitectures: list[str] | None = Noneoutput_hidden_states: bool | None = Falsereturn_dict: bool | None = Truedtype: typing.Union[str, ForwardRef('torch.dtype'), NoneType] = Nonechunk_size_feed_forward: int = 0is_encoder_decoder: bool = Falseid2label: dict[int, str] | dict[str, str] | None = Nonelabel2id: dict[str, int] | dict[str, str] | None = Noneproblem_type: typing.Optional[typing.Literal['regression', 'single_label_classification', 'multi_label_classification']] = Nonehidden_size: int = 1152intermediate_size: int = 4304num_hidden_layers: int = 27num_attention_heads: int = 16num_channels: int = 3patch_size: int | list[int] | tuple[int, int] = 16hidden_act: str = 'gelu_pytorch_tanh'layer_norm_eps: float = 1e-06attention_dropout: float | int = 0.0num_patches: int = 256spatial_merge_size: int = 2 )

Parameters

  • hidden_size (int, optional, defaults to 1152) — Dimension of the hidden representations.
  • intermediate_size (int, optional, defaults to 4304) — Dimension of the MLP representations.
  • num_hidden_layers (int, optional, defaults to 27) — Number of hidden layers in the Transformer decoder.
  • num_attention_heads (int, optional, defaults to 16) — Number of attention heads for each attention layer in the Transformer decoder.
  • num_channels (int, optional, defaults to 3) — The number of input channels.
  • patch_size (Union[int, list[int], tuple[int, int]], optional, defaults to 16) — The size (resolution) of each patch.
  • hidden_act (str, optional, defaults to gelu_pytorch_tanh) — The non-linear activation function (function or string) in the decoder. For example, "gelu", "relu", "silu", etc.
  • layer_norm_eps (float, optional, defaults to 1e-06) — The epsilon used by the layer normalization layers.
  • attention_dropout (Union[float, int], optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
  • num_patches (int, optional, defaults to 256) — Number of patches in the learned reference positional-embedding grid.
  • spatial_merge_size (int, optional, defaults to 2) — The size of the spatial merge window used to reduce the number of visual tokens by merging neighboring patches.

This is the configuration class to store the configuration of a Cosmos3EdgeModel. It is used to instantiate a Cosmos3 Edge model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the nvidia/Cosmos3-Edge-Reasoner

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Cosmos3EdgeProcessor

class transformers.Cosmos3EdgeProcessor

< >

( image_processor = Nonetokenizer = Nonevideo_processor = Nonechat_template = None**kwargs )

Parameters

  • image_processor (Cosmos3EdgeImageProcessor) — The image processor is a required input.
  • tokenizer (tokenizer_class) — The tokenizer is a required input.
  • video_processor (Cosmos3EdgeVideoProcessor) — The video processor is a required input.
  • chat_template (str) — A Jinja template to convert lists of messages in a chat into a tokenizable string.

Constructs a Cosmos3EdgeProcessor which wraps a image processor, a tokenizer, and a video processor into a single processor.

Cosmos3EdgeProcessor offers all the functionalities of Cosmos3EdgeImageProcessor, tokenizer_class, and Cosmos3EdgeVideoProcessor. See the ~Cosmos3EdgeImageProcessor, ~tokenizer_class, and ~Cosmos3EdgeVideoProcessor for more information.

__call__

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor'], NoneType] = Nonetext: str | list[str] | list[list[str]] | None = Nonevideos: typing.Union[list['PIL.Image.Image'], numpy.ndarray, ForwardRef('torch.Tensor'), list[numpy.ndarray], list['torch.Tensor'], list[list['PIL.Image.Image']], list[list[numpy.ndarray]], list[list['torch.Tensor']], transformers.video_utils.URL, list[transformers.video_utils.URL], list[list[transformers.video_utils.URL]], transformers.video_utils.Path, list[transformers.video_utils.Path], list[list[transformers.video_utils.Path]], NoneType] = Noneaudio: typing.Union[numpy.ndarray, ForwardRef('torch.Tensor'), collections.abc.Sequence[numpy.ndarray], collections.abc.Sequence['torch.Tensor'], NoneType] = None**kwargs: Unpack )

Parameters

  • images (Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]], optional) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set do_rescale=False.
  • text (Union[str, list[str], list[list[str]]], optional) — The sequence or batch of sequences to be encoded. Each sequence can be a string or a list of strings (pretokenized string). If you pass a pretokenized input, set is_split_into_words=True to avoid ambiguity with batched inputs.
  • videos (Union[list[PIL.Image.Image], numpy.ndarray, torch.Tensor, list[numpy.ndarray], list[torch.Tensor], list[list[PIL.Image.Image]], list[list[numpy.ndarray]], list[list[torch.Tensor]], ~video_utils.URL, list[~video_utils.URL], list[list[~video_utils.URL]], ~video_utils.Path, list[~video_utils.Path], list[list[~video_utils.Path]]], optional) — Video to preprocess. Expects a single or batch of videos with pixel values ranging from 0 to 255. If passing in videos with pixel values between 0 and 1, set do_rescale=False.
  • audio (Union[numpy.ndarray, torch.Tensor, collections.abc.Sequence[numpy.ndarray], collections.abc.Sequence[torch.Tensor]], optional) — The audio or batch of audios to be prepared. Each audio can be a NumPy array or PyTorch tensor. In case of a NumPy array/PyTorch tensor, each audio should be of shape (C, T), where C is a number of channels, and T is the sample length of the audio.
  • return_tensors (str or TensorType, optional) — If set, will return tensors of a particular framework. Acceptable values are:

    • 'pt': Return PyTorch torch.Tensor objects.
    • 'np': Return NumPy np.ndarray objects.
  • **kwargs (ProcessingKwargs, optional) — Additional processing options for each modality (text, images, videos, audio). Model-specific parameters are listed above; see the TypedDict class for the complete list of supported arguments.

apply_chat_template

< >

( conversation: list[dict[str, str]] | list[list[dict[str, str]]]chat_template: str | None = Nonetools: list[dict] | None = Nonedocuments: list[dict[str, str]] | None = Noneadd_generation_prompt: bool = Falsecontinue_final_message: bool | str = Falsereturn_assistant_tokens_mask: bool = Falsetokenize: bool = Falsereturn_tensors: str | transformers.utils.generic.TensorType | None = Nonereturn_dict: bool = Falseload_audio_from_video: bool = Falseprocessor_kwargs: dict | None = None**kwargs )

Parameters

  • conversation (Union[list[Dict, [str, str]], list[list[dict[str, str]]]]) — The conversation to format.
  • chat_template (Optional[str], optional) — The Jinja template to use for formatting the conversation. If not provided, the tokenizer’s chat template is used.

Similar to the apply_chat_template method on tokenizers, this method applies a Jinja template to input conversations to turn them into a single tokenizable string.

The input is expected to be in the following format, where each message content is a list consisting of text and optionally image or video inputs. One can also provide an image, video, URL or local path which will be used to form pixel_values when return_dict=True. If not provided, one will get only the formatted text, optionally tokenized text.

conversation = [ { “role”: “user”, “content”: [ {“type”: “image”, “url”: “https://www.ilankelman.org/stopsigns/australia.jpg”}, {“type”: “text”, “text”: “Please describe this image in detail.”}, ], }, ]

Cosmos3EdgeImageProcessor

class transformers.Cosmos3EdgeImageProcessor

< >

( **kwargs: Unpack )

Parameters

  • patch_size (int, kwargs, optional, defaults to 16) — Spatial patch size of the vision encoder.
  • merge_size (int, kwargs, optional, defaults to 2) — Number of adjacent patches merged along each spatial axis by the projector.
  • **kwargs (ImagesKwargs, optional) — Additional image preprocessing options. Model-specific kwargs are listed above; see the TypedDict class for the complete list of supported arguments.

Constructs a Cosmos3EdgeImageProcessor image processor.

preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]**kwargs: Unpack ) → ~image_processing_base.BatchFeature

Parameters

  • images (Union[PIL.Image.Image, numpy.ndarray, torch.Tensor, list[PIL.Image.Image], list[numpy.ndarray], list[torch.Tensor]]) — Image to preprocess. Expects a single or batch of images with pixel values ranging from 0 to 255. If passing in images with pixel values between 0 and 1, set do_rescale=False.
  • patch_size (int, kwargs, optional, defaults to 16) — Spatial patch size of the vision encoder.
  • merge_size (int, kwargs, optional, defaults to 2) — Number of adjacent patches merged along each spatial axis by the projector.
  • return_tensors (str or TensorType, optional) — Returns stacked tensors if set to 'pt', otherwise returns a list of tensors.
  • **kwargs (ImagesKwargs, optional) — Additional image preprocessing options. Model-specific kwargs are listed above; see the TypedDict class for the complete list of supported arguments.

Returns

~image_processing_base.BatchFeature

  • data (dict) — Dictionary of lists/arrays/tensors returned by the call method (‘pixel_values’, etc.).
  • tensor_type (Union[None, str, TensorType], optional) — You can give a tensor_type here to convert the lists of integers in PyTorch/Numpy Tensors at initialization.

Cosmos3EdgeImageProcessorPil

class transformers.Cosmos3EdgeImageProcessorPil

< >

( **kwargs: Unpack )

Dynamically resize images and return packed, unpadded SigLIP2 patches.

patch_size (int, optional, defaults to 16): Spatial patch size of the vision encoder. merge_size (int, optional, defaults to 2): Number of adjacent patches merged along each spatial axis by the projector.

preprocess

< >

( images: typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]**kwargs: Unpack )

Cosmos3EdgeVideoProcessor

class transformers.Cosmos3EdgeVideoProcessor

< >

( **kwargs: Unpack )

Parameters

  • do_resize (bool, optional, defaults to self.do_resize) — Whether to resize the video’s (height, width) dimensions to the specified size. Can be overridden by the do_resize parameter in the preprocess method.
  • size (dict, optional, defaults to self.size) — Size of the output video after resizing. Can be overridden by the size parameter in the preprocess method.
  • size_divisor (int, optional, defaults to self.size_divisor) — The size by which to make sure both the height and width can be divided.
  • default_to_square (bool, optional, defaults to self.default_to_square) — Whether to default to a square video when resizing, if size is an int.
  • resample (PILImageResampling, optional, defaults to self.resample) — Resampling filter to use if resizing the video. Only has an effect if do_resize is set to True. Can be overridden by the resample parameter in the preprocess method.
  • do_center_crop (bool, optional, defaults to self.do_center_crop) — Whether to center crop the video to the specified crop_size. Can be overridden by do_center_crop in the preprocess method.
  • crop_size (dict[str, int] optional, defaults to self.crop_size) — Size of the output video after applying center_crop. Can be overridden by crop_size in the preprocess method.
  • do_rescale (bool, optional, defaults to self.do_rescale) — Whether to rescale the video by the specified scale rescale_factor. Can be overridden by the do_rescale parameter in the preprocess method.
  • rescale_factor (int or float, optional, defaults to self.rescale_factor) — Scale factor to use if rescaling the video. Only has an effect if do_rescale is set to True. Can be overridden by the rescale_factor parameter in the preprocess method.
  • do_normalize (bool, optional, defaults to self.do_normalize) — Whether to normalize the video. Can be overridden by the do_normalize parameter in the preprocess method. Can be overridden by the do_normalize parameter in the preprocess method.
  • image_mean (float or list[float], optional, defaults to self.image_mean) — Mean to use if normalizing the video. This is a float or list of floats the length of the number of channels in the video. Can be overridden by the image_mean parameter in the preprocess method. Can be overridden by the image_mean parameter in the preprocess method.
  • image_std (float or list[float], optional, defaults to self.image_std) — Standard deviation to use if normalizing the video. This is a float or list of floats the length of the number of channels in the video. Can be overridden by the image_std parameter in the preprocess method. Can be overridden by the image_std parameter in the preprocess method.
  • do_convert_rgb (bool, optional, defaults to self.image_std) — Whether to convert the video to RGB.
  • video_metadata (VideoMetadata, optional) — Metadata of the video containing information about total duration, fps and total number of frames.
  • do_sample_frames (int, optional, defaults to self.do_sample_frames) — Whether to sample frames from the video before processing or to process the whole video.
  • num_frames (int, optional, defaults to self.num_frames) — Maximum number of frames to sample when do_sample_frames=True.
  • fps (int or float, optional, defaults to self.fps) — Target frames to sample per second when do_sample_frames=True.
  • return_tensors (str or TensorType, optional) — Returns stacked tensors if set to `pt, otherwise returns a list of tensors.
  • data_format (ChannelDimension or str, optional, defaults to ChannelDimension.FIRST) — The channel dimension format for the output video. Can be one of:

    • "channels_first" or ChannelDimension.FIRST: video in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: video in (height, width, num_channels) format.
    • Unset: Use the channel dimension format of the input video.
  • input_data_format (ChannelDimension or str, optional) — The channel dimension format for the input video. If unset, the channel dimension format is inferred from the input video. Can be one of:

    • "channels_first" or ChannelDimension.FIRST: video in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: video in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: video in (height, width) format.
  • device (torch.device, optional) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_metadata (bool, optional) — Whether to return video metadata or not.
  • patch_size (int, optional, defaults to 16) — Spatial patch size of the vision encoder.
  • temporal_patch_size (int, optional, defaults to 1) — Temporal patch size of the vision encoder. Cosmos3 Edge processes each sampled frame independently.
  • merge_size (int, optional, defaults to 2) — Spatial merge size applied by the vision projector.

Constructs a video processor that dynamically resizes and packs Cosmos3 Edge video frames.

preprocess

< >

( videos: typing.Union[list['PIL.Image.Image'], numpy.ndarray, ForwardRef('torch.Tensor'), list[numpy.ndarray], list['torch.Tensor'], list[list['PIL.Image.Image']], list[list[numpy.ndarray]], list[list['torch.Tensor']], transformers.video_utils.URL, list[transformers.video_utils.URL], list[list[transformers.video_utils.URL]], transformers.video_utils.Path, list[transformers.video_utils.Path], list[list[transformers.video_utils.Path]]]**kwargs: Unpack )

Parameters

  • do_resize (bool, optional, defaults to self.do_resize) — Whether to resize the video’s (height, width) dimensions to the specified size. Can be overridden by the do_resize parameter in the preprocess method.
  • size (dict, optional, defaults to self.size) — Size of the output video after resizing. Can be overridden by the size parameter in the preprocess method.
  • size_divisor (int, optional, defaults to self.size_divisor) — The size by which to make sure both the height and width can be divided.
  • default_to_square (bool, optional, defaults to self.default_to_square) — Whether to default to a square video when resizing, if size is an int.
  • resample (PILImageResampling, optional, defaults to self.resample) — Resampling filter to use if resizing the video. Only has an effect if do_resize is set to True. Can be overridden by the resample parameter in the preprocess method.
  • do_center_crop (bool, optional, defaults to self.do_center_crop) — Whether to center crop the video to the specified crop_size. Can be overridden by do_center_crop in the preprocess method.
  • crop_size (dict[str, int] optional, defaults to self.crop_size) — Size of the output video after applying center_crop. Can be overridden by crop_size in the preprocess method.
  • do_rescale (bool, optional, defaults to self.do_rescale) — Whether to rescale the video by the specified scale rescale_factor. Can be overridden by the do_rescale parameter in the preprocess method.
  • rescale_factor (int or float, optional, defaults to self.rescale_factor) — Scale factor to use if rescaling the video. Only has an effect if do_rescale is set to True. Can be overridden by the rescale_factor parameter in the preprocess method.
  • do_normalize (bool, optional, defaults to self.do_normalize) — Whether to normalize the video. Can be overridden by the do_normalize parameter in the preprocess method. Can be overridden by the do_normalize parameter in the preprocess method.
  • image_mean (float or list[float], optional, defaults to self.image_mean) — Mean to use if normalizing the video. This is a float or list of floats the length of the number of channels in the video. Can be overridden by the image_mean parameter in the preprocess method. Can be overridden by the image_mean parameter in the preprocess method.
  • image_std (float or list[float], optional, defaults to self.image_std) — Standard deviation to use if normalizing the video. This is a float or list of floats the length of the number of channels in the video. Can be overridden by the image_std parameter in the preprocess method. Can be overridden by the image_std parameter in the preprocess method.
  • do_convert_rgb (bool, optional, defaults to self.image_std) — Whether to convert the video to RGB.
  • video_metadata (VideoMetadata, optional) — Metadata of the video containing information about total duration, fps and total number of frames.
  • do_sample_frames (int, optional, defaults to self.do_sample_frames) — Whether to sample frames from the video before processing or to process the whole video.
  • num_frames (int, optional, defaults to self.num_frames) — Maximum number of frames to sample when do_sample_frames=True.
  • fps (int or float, optional, defaults to self.fps) — Target frames to sample per second when do_sample_frames=True.
  • return_tensors (str or TensorType, optional) — Returns stacked tensors if set to `pt, otherwise returns a list of tensors.
  • data_format (ChannelDimension or str, optional, defaults to ChannelDimension.FIRST) — The channel dimension format for the output video. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: video in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: video in (height, width, num_channels) format.
    • Unset: Use the channel dimension format of the input video.
  • input_data_format (ChannelDimension or str, optional) — The channel dimension format for the input video. If unset, the channel dimension format is inferred from the input video. Can be one of:
    • "channels_first" or ChannelDimension.FIRST: video in (num_channels, height, width) format.
    • "channels_last" or ChannelDimension.LAST: video in (height, width, num_channels) format.
    • "none" or ChannelDimension.NONE: video in (height, width) format.
  • device (torch.device, optional) — The device to process the videos on. If unset, the device is inferred from the input videos.
  • return_metadata (bool, optional) — Whether to return video metadata or not.

Cosmos3EdgeModel

class transformers.Cosmos3EdgeModel

< >

( config: Cosmos3EdgeConfig )

Parameters

  • config (Cosmos3EdgeConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare Cosmos3 Edge Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Noneuse_cache: bool | None = Nonepixel_values: typing.Optional[torch.Tensor] = Nonepixel_values_videos: typing.Optional[torch.FloatTensor] = Noneimage_grid_thw: typing.Optional[torch.LongTensor] = Nonevideo_grid_thw: typing.Optional[torch.LongTensor] = Nonemm_token_type_ids: typing.Optional[torch.IntTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • pixel_values (torch.Tensor of shape (batch_size, num_channels, image_size, image_size), optional) — The tensors corresponding to the input images. Pixel values can be obtained using Cosmos3EdgeImageProcessor. See Cosmos3EdgeImageProcessor.__call__() for details (Cosmos3EdgeProcessor uses Cosmos3EdgeImageProcessor for processing images).
  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_frames, num_channels, frame_size, frame_size), optional) — The tensors corresponding to the input video. Pixel values for videos can be obtained using Cosmos3EdgeVideoProcessor. See Cosmos3EdgeVideoProcessor.__call__() for details (Cosmos3EdgeProcessor uses Cosmos3EdgeVideoProcessor for processing videos).
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height, and width of the feature grid for each image.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height, and width of the feature grid for each video.
  • mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens matching each modality. For example text (0), image (1), video (2). Multimodal token type ids can be obtained using AutoProcessor. See ProcessorMixin.call() for details.

Returns

BaseModelOutputWithPast or tuple(torch.FloatTensor)

A BaseModelOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

The Cosmos3EdgeModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

    If past_key_values is used only the last hidden-state of the sequences of shape (batch_size, 1, hidden_size) is output.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if config.is_encoder_decoder=True in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

get_image_features

< >

( pixel_values: FloatTensorimage_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (num_patches, num_channels * patch_size * patch_size)) — Packed image patches.
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height, and width dimensions of every packed image patch grid.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

get_video_features

< >

( pixel_values_videos: FloatTensorvideo_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values_videos (torch.FloatTensor of shape (num_patches, num_channels * patch_size * patch_size)) — Packed video-frame patches.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height, and width dimensions of every packed video patch grids.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Cosmos3EdgeTextModel

class transformers.Cosmos3EdgeTextModel

< >

( config: Cosmos3EdgeTextConfig )

Parameters

  • config (Cosmos3EdgeTextConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare Cosmos3 Edge Text Model outputting raw hidden-states without any specific head on to.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Noneuse_cache: bool | None = None**kwargs ) → BaseModelOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).

Returns

BaseModelOutputWithPast or tuple(torch.FloatTensor)

A BaseModelOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

The Cosmos3EdgeTextModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

    If past_key_values is used only the last hidden-state of the sequences of shape (batch_size, 1, hidden_size) is output.

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks and optionally if config.is_encoder_decoder=True in the cross-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Cosmos3EdgeVisionModel

class transformers.Cosmos3EdgeVisionModel

< >

( config: Cosmos3EdgeVisionConfig )

Packed variable-resolution SigLIP2 vision tower used by Cosmos3 Edge.

forward

< >

( pixel_values: FloatTensorgrid_thw: LongTensor**kwargs ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input images. Pixel values can be obtained using Cosmos3EdgeImageProcessor. See Cosmos3EdgeImageProcessor.__call__() for details (Cosmos3EdgeProcessor uses Cosmos3EdgeImageProcessor for processing images).
  • grid_thw (torch.LongTensor of shape (num_images_or_videos, 3)) — The temporal, height, and width patch-grid dimensions for each packed image or video.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

The Cosmos3EdgeVisionModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Cosmos3EdgeForConditionalGeneration

class transformers.Cosmos3EdgeForConditionalGeneration

< >

( config )

forward

< >

( input_ids: typing.Optional[torch.LongTensor] = Noneattention_mask: typing.Optional[torch.Tensor] = Noneposition_ids: typing.Optional[torch.LongTensor] = Nonepast_key_values: transformers.cache_utils.Cache | None = Noneinputs_embeds: typing.Optional[torch.FloatTensor] = Nonelabels: typing.Optional[torch.LongTensor] = Noneuse_cache: bool | None = Nonepixel_values: typing.Optional[torch.Tensor] = Nonepixel_values_videos: typing.Optional[torch.FloatTensor] = Noneimage_grid_thw: typing.Optional[torch.LongTensor] = Nonevideo_grid_thw: typing.Optional[torch.LongTensor] = Nonemm_token_type_ids: typing.Optional[torch.IntTensor] = Nonelogits_to_keep: typing.Union[int, torch.Tensor] = 0**kwargs: Unpack ) → CausalLMOutputWithPast or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • attention_mask (torch.Tensor of shape (batch_size, sequence_length), optional) — Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

    • 1 for tokens that are not masked,
    • 0 for tokens that are masked.

    What are attention masks?

  • position_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) — Indices of positions of each input sequence tokens in the position embeddings. Selected in the range [0, config.n_positions - 1].

    What are position IDs?

  • past_key_values (~cache_utils.Cache, optional) — Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention blocks) that can be used to speed up sequential decoding. This typically consists in the past_key_values returned by the model at a previous stage of decoding, when use_cache=True or config.use_cache=True.

    Only Cache instance is allowed as input, see our kv cache guide. If no past_key_values are passed, DynamicCache will be initialized by default.

    The model will output the same cache format that is fed as input.

    If past_key_values are used, the user is expected to input only unprocessed input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, unprocessed_length) instead of all input_ids of shape (batch_size, sequence_length).

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • labels (torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for computing the masked language modeling loss. Indices should either be in [0, ..., config.vocab_size] or -100 (see input_ids docstring). Tokens with indices set to -100 are ignored (masked), and the loss is only computed for tokens with labels in [0, ..., config.vocab_size].
  • use_cache (bool, optional) — If set to True, past_key_values key value states are returned and can be used to speed up decoding (see past_key_values).
  • pixel_values (torch.Tensor of shape (batch_size, num_channels, image_size, image_size), optional) — The tensors corresponding to the input images. Pixel values can be obtained using Cosmos3EdgeImageProcessor. See Cosmos3EdgeImageProcessor.__call__() for details (Cosmos3EdgeProcessor uses Cosmos3EdgeImageProcessor for processing images).
  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_frames, num_channels, frame_size, frame_size), optional) — The tensors corresponding to the input video. Pixel values for videos can be obtained using Cosmos3EdgeVideoProcessor. See Cosmos3EdgeVideoProcessor.__call__() for details (Cosmos3EdgeProcessor uses Cosmos3EdgeVideoProcessor for processing videos).
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height, and width of the feature grid for each image.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height, and width of the feature grid for each video.
  • mm_token_type_ids (torch.IntTensor of shape (batch_size, sequence_length), optional) — Indices of input sequence tokens matching each modality. For example text (0), image (1), video (2). Multimodal token type ids can be obtained using AutoProcessor. See ProcessorMixin.call() for details.
  • logits_to_keep (Union[int, torch.Tensor], optional, defaults to 0) — If an int, compute logits for the last logits_to_keep tokens. If 0, calculate logits for all input_ids (special case). Only last token logits are needed for generation, and calculating them only for that token can save memory, which becomes pretty significant for long sequences or large vocabulary size. If a torch.Tensor, must be 1D corresponding to the indices to keep in the sequence length dimension. This is useful when using packed tensor format (single dimension for batch and sequence length).

Returns

CausalLMOutputWithPast or tuple(torch.FloatTensor)

A CausalLMOutputWithPast or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

The Cosmos3EdgeForConditionalGeneration forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

  • loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).

  • logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

  • past_key_values (Cache, optional, returned when use_cache=True is passed or when config.use_cache=True) — It is a Cache instance. For more details, see our kv cache guide.

    Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see past_key_values input) to speed up sequential decoding.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Example:

>>> from transformers import AutoProcessor, Cosmos3EdgeForConditionalGeneration

>>> model = Cosmos3EdgeForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
>>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")

>>> messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
            },
            {"type": "text", "text": "Describe the image."},
        ],
    }
]

>>> inputs = processor.apply_chat_template(
    messages,
    tokenize=True,
    add_generation_prompt=True,
    return_dict=True,
    return_tensors="pt"
)

>>> # Generate
>>> generated_ids = model.generate(**inputs, max_new_tokens=1024)
>>> generated_ids_trimmed = [out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]
>>> output_text = processor.batch_decode(generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
>>> print(output_text)

get_image_features

< >

( pixel_values: FloatTensorimage_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input images.
  • image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Example:

>>> from PIL import Image
>>> from transformers import AutoProcessor, Cosmos3EdgeForConditionalGeneration

>>> model = Cosmos3EdgeForConditionalGeneration.from_pretrained("nvidia/Cosmos3-Edge-Reasoner")
>>> processor = AutoProcessor.from_pretrained("nvidia/Cosmos3-Edge-Reasoner")

>>> messages = [
...     {
...         "role": "user", "content": [
...             {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
...             {"type": "text", "text": "Where is the cat standing?"},
...         ]
...     },
... ]

>>> inputs = processor.apply_chat_template(
...     messages,
...     tokenize=True,
...     return_dict=True,
...     return_tensors="pt",
...     add_generation_prompt=True
... )
>>> # Generate
>>> generate_ids = model.generate(**inputs)
>>> processor.batch_decode(generate_ids, skip_special_tokens=True)[0]

get_video_features

< >

( pixel_values_videos: FloatTensorvideo_grid_thw: typing.Optional[torch.LongTensor] = None**kwargs: Unpack ) → BaseModelOutputWithPooling or tuple(torch.FloatTensor)

Parameters

  • pixel_values_videos (torch.FloatTensor of shape (batch_size, num_channels, image_size, image_size)) — The tensors corresponding to the input videos.
  • video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.

Returns

BaseModelOutputWithPooling or tuple(torch.FloatTensor)

A BaseModelOutputWithPooling or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (Cosmos3EdgeConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • pooler_output (torch.FloatTensor of shape (batch_size, hidden_size)) — Last layer hidden-state of the first token of the sequence (classification token) after further processing through the layers used for the auxiliary pretraining task. E.g. for BERT-family of models, this returns the classification token after processing through a linear layer and a tanh activation function. The linear layer weights are trained from the next sentence prediction (classification) objective during pretraining.

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

  • attentions (tuple(torch.FloatTensor), optional, returned when output_attentions=True is passed or when config.output_attentions=True) — Tuple of torch.FloatTensor (one for each layer) of shape (batch_size, num_heads, sequence_length, sequence_length).

    Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.

Example:

>>> from PIL import Image
>>> from transformers import AutoProcessor, Cosmos3EdgeForConditionalGeneration

>>> model = Cosmos3EdgeForConditionalGeneration.from_pretrained("nvidia/Cosmos3-Edge-Reasoner")
>>> processor = AutoProcessor.from_pretrained("nvidia/Cosmos3-Edge-Reasoner")

>>> messages = [
...     {
...         "role": "user", "content": [
...             {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
...             {"type": "text", "text": "Where is the cat standing?"},
...         ]
...     },
... ]

>>> inputs = processor.apply_chat_template(
...     messages,
...     tokenize=True,
...     return_dict=True,
...     return_tensors="pt",
...     add_generation_prompt=True
... )
>>> # Generate
>>> generate_ids = model.generate(**inputs)
>>> processor.batch_decode(generate_ids, skip_special_tokens=True)[0]
Update on GitHub

←ColQwen2 Cosmos3 Omni→