# 소형 비전 언어 모델(SmolVLM)[[smolvlm]] ## 개요[[overview]] [SmolVLM2](https://huggingface.co/papers/2504.05299) ([블로그 글](https://huggingface.co/blog/smolvlm2)) 은 Idefics3 모델을 개선한 버전으로, 두 가지 주요 차이점이 있습니다: - 텍스트 모델로 SmolLM2를 사용합니다. - 한 장의 이미지뿐 아니라 여러 장의 이미지와 비디오 입력도 지원합니다. ## 사용 팁[[usage-tips]] 입력된 이미지는 설정에 따라 원본 해상도를 유지하거나 크기를 조절할 수 있습니다. 이때 이미지 크기 조절 여부와 방식은 `do_resize`와 `size` 파라미터로 결정됩니다. 비디오의 경우에는 업샘플링을 하면 안 됩니다. 만약 `do_resize`가 `True`일 경우, 모델은 기본적으로 이미지의 가장 긴 변을 4*512 픽셀이 되도록 크기를 조절합니다. 이 기본 동작은 `size` 파라미터에 딕셔너리를 전달하여 원하는 값으로 직접 설정할 수 있습니다. 예를 들어, 기본값은 `{"longest_edge": 4 * 512}` 이여도 사용자 필요에 따라 다른 값으로 변경할 수 있습니다. 다음은 리사이징을 제어하고 사용자 정의 크기로 변경하는 방법입니다: ```python image_processor = SmolVLMImageProcessor(do_resize=True, size={"longest_edge": 2 * 512}, max_image_size=512) ``` 또한, `max_image_size` 매개변수는 이미지를 분할하는 정사각형 패치의 크기를 제어합니다. 이 값은 기본적으로 512로 설정되어 있으며 필요에 따라 조정 가능합니다. 이미지 처리기는 리사이징을 마친 후, `max_image_size` 값을 기준으로 이미지를 여러 개의 정사각형 패치로 분할합니다. 이 모델의 기여자는 [orrzohar](https://huggingface.co/orrzohar) 입니다. ## 사용 예시[[usage-example]] ### 단일 미디어 추론[[single-media-inference]] 이 모델은 이미지와 비디오를 모두 입력으로 받을 수 있지만, 한 번에 사용할 수 있는 미디어는 반드시 하나의 종류여야 합니다. 관련 예시 코드는 다음과 같습니다. ```python import torch from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM2-256M-Video-Instruct") model = AutoModelForImageTextToText.from_pretrained( "HuggingFaceTB/SmolVLM2-256M-Video-Instruct", dtype=torch.bfloat16, device_map="auto" ) conversation = [ { "role": "user", "content":[ {"type": "image", "url": "http://images.cocodataset.org/val2017/000000039769.jpg"}, {"type": "text", "text": "이 이미지에 대해 설명해주세요."} ] } ] inputs = processor.apply_chat_template( conversation, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device, dtype=torch.bfloat16) output_ids = model.generate(**inputs, max_new_tokens=128) generated_texts = processor.batch_decode(output_ids, skip_special_tokens=True) print(generated_texts) # Video conversation = [ { "role": "user", "content": [ {"type": "video", "path": "/path/to/video.mp4"}, {"type": "text", "text": "이 비디오에 대해 자세히 설명해주세요."} ] }, ] inputs = processor.apply_chat_template( conversation, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device, dtype=torch.bfloat16) generated_ids = model.generate(**inputs, do_sample=False, max_new_tokens=100) generated_texts = processor.batch_decode(generated_ids, skip_special_tokens=True) print(generated_texts[0]) ``` ### 배치 다중 미디어 추론[[batch-mixed-media-inference]] 이 모델은 여러 이미지, 비디오, 텍스트로 구성된 입력을 한 번에 배치 형태로 처리할 수 있습니다. 관련 예시는 다음과 같습니다. ```python import torch from transformers import AutoProcessor, AutoModelForImageTextToText processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM2-256M-Video-Instruct") model = AutoModelForImageTextToText.from_pretrained( "HuggingFaceTB/SmolVLM2-256M-Video-Instruct", dtype=torch.bfloat16, device_map="auto" ) # 첫 번째 이미지에 대한 구성 conversation1 = [ { "role": "user", "content": [ {"type": "image", "path": "/path/to/image.jpg"}, {"type": "text", "text": "이 이미지에 대해 설명해주세요."} ] } ] # 두 장의 이미지를 포함한 구성 conversation2 = [ { "role": "user", "content": [ {"type": "image", "path": "/path/to/image.jpg"}, {"type": "image", "path": "/path/to/image.jpg"}, {"type": "text", "text": "그림에 무엇이 적혀있나요?"} ] } ] # 텍스트만 포함하고 있는 구성 conversation3 = [ {"role": "user","content": "당신은 누구인가요?"} ] conversations = [conversation1, conversation2, conversation3] inputs = processor.apply_chat_template( conversation, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device, dtype=torch.bfloat16) generated_ids = model.generate(**inputs, do_sample=False, max_new_tokens=100) generated_texts = processor.batch_decode(generated_ids, skip_special_tokens=True) print(generated_texts[0]) ``` ## SmolVLMConfig[[transformers.SmolVLMConfig]][[transformers.SmolVLMConfig]] #### transformers.SmolVLMConfig[[transformers.SmolVLMConfig]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/configuration_smolvlm.py#L115) This is the configuration class to store the configuration of a [SmolVLMModel](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMModel). It is used to instantiate a SmolVLM 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 model of the SmolVLM [HuggingFaceTB/SmolVLM2-2.2B-Instruct](https://huggingface.co/HuggingFaceTB/SmolVLM2-2.2B-Instruct) architecture. Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.0.0rc2/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the documentation from [PreTrainedConfig](/docs/transformers/v5.0.0rc2/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information. Example: ```python >>> from transformers import SmolVLMModel, SmolVLMConfig >>> # Initializing configuration >>> configuration = SmolVLMConfig() >>> # Initializing a model from the configuration >>> model = SmolVLMModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` **Parameters:** use_cache (`bool`, *optional*, defaults to `True`) : Whether or not the model should cache the key/value pairs of the attention mechanism. Only relevant if `config.is_decoder=True`. image_token_id (`int`, *optional*, defaults to 128257) : The id of the "image" token. tie_word_embeddings (`bool`, *optional*, defaults to `False`) : Whether or not to tie the word embeddings with the token embeddings. vision_config (`IdeficsVisionConfig` or `dict`, *optional*, defaults to `IdeficsVisionConfig`) : Custom vision config or dict for the vision tower text_config (`PreTrainedConfig` or `dict`, *optional*, defaults to `LlamaConfig`) : Custom text config or dict for the text model scale_factor (`int`, *optional*, defaults to 2) : The scale factor for the image encoder. pad_token_id (`int`, *optional*, defaults to 128002) : The id of the padding token. ## SmolVLMVisionConfig[[transformers.SmolVLMVisionConfig]][[transformers.SmolVLMVisionConfig]] #### transformers.SmolVLMVisionConfig[[transformers.SmolVLMVisionConfig]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/configuration_smolvlm.py#L30) This is the configuration class to store the configuration of a `SmolVLMVisionModel`. It is used to instantiate a SmolVLM vision encoder according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the SigLIP checkpoint [google/siglip-so400m-patch14-384](https://huggingface.co/google/siglip-so400m-patch14-384) used in SmolVLM [HuggingFaceTB/SmolVLM2-2.2B-Instruct](https://huggingface.co/HuggingFaceTB/SmolVLM2-2.2B-Instruct). Configuration objects inherit from [PreTrainedConfig](/docs/transformers/v5.0.0rc2/ko/main_classes/configuration#transformers.PreTrainedConfig) and can be used to control the model outputs. Read the documentation from [PreTrainedConfig](/docs/transformers/v5.0.0rc2/ko/main_classes/configuration#transformers.PreTrainedConfig) for more information. Example: ```python >>> from transformers.models.smolvlm.modeling_smolvlm import SmolVLMVisionTransformer >>> from transformers.models.smolvlm.configuration_smolvlm import SmolVLMVisionConfig >>> # Initializing a SmolVLMVisionConfig with google/siglip-so400m-patch14-384 style configuration >>> configuration = SmolVLMVisionConfig() >>> # Initializing a SmolVLMVisionTransformer (with random weights) from the google/siglip-so400m-patch14-384 style configuration >>> model = SmolVLMVisionTransformer(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` **Parameters:** hidden_size (`int`, *optional*, defaults to 1152) : Dimensionality of the encoder layers and the pooler layer. intermediate_size (`int`, *optional*, defaults to 3072) : Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder. num_hidden_layers (`int`, *optional*, defaults to 12) : Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 16) : Number of attention heads for each attention layer in the Transformer encoder. num_channels (`int`, *optional*, defaults to 3) : Number of channels in the input images. image_size (`int`, *optional*, defaults to 224) : The size (resolution) of each image. patch_size (`int`, *optional*, defaults to 32) : The size (resolution) of each patch. hidden_act (`str` or `function`, *optional*, defaults to `"gelu_pytorch_tanh"`) : The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"selu"` and `"gelu_new"` `"quick_gelu"` are supported. layer_norm_eps (`float`, *optional*, defaults to 1e-06) : The epsilon used by the layer normalization layers. attention_dropout (`float`, *optional*, defaults to 0.0) : The dropout ratio for the attention probabilities. initializer_range (`float`, *optional*, defaults to 0.02) : The standard deviation of the truncated_normal_initializer for initializing all weight matrices. ## Idefics3VisionTransformer[[transformers.SmolVLMVisionTransformer]][[transformers.SmolVLMVisionTransformer]] #### transformers.SmolVLMVisionTransformer[[transformers.SmolVLMVisionTransformer]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/modeling_smolvlm.py#L313) The SmolVLM Vision Transformer Model outputting raw image embedding. This model inherits from [PreTrainedModel](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.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](https://pytorch.org/docs/stable/nn.html#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. **Parameters:** config ([SmolVLMVisionConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMVisionConfig)) : 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()](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. ## SmolVLMModel[[transformers.SmolVLMModel]][[transformers.SmolVLMModel]] #### transformers.SmolVLMModel[[transformers.SmolVLMModel]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/modeling_smolvlm.py#L453) SmolVLM model consisting of a SIGLIP vision encoder and Llama3 language decoder This model inherits from [PreTrainedModel](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.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](https://pytorch.org/docs/stable/nn.html#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. forwardtransformers.SmolVLMModel.forwardhttps://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/modeling_smolvlm.py#L571[{"name": "input_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "past_key_values", "val": ": typing.Optional[transformers.cache_utils.Cache] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "pixel_values", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "pixel_attention_mask", "val": ": typing.Optional[torch.BoolTensor] = None"}, {"name": "image_hidden_states", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "use_cache", "val": ": typing.Optional[bool] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}, {"name": "cache_position", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.modeling_flash_attention_utils.FlashAttentionKwargs]"}]- **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](/docs/transformers/v5.0.0rc2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.0.0rc2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.0.0rc2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#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?](../glossary#attention-mask) - **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?](../glossary#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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance is allowed as input, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). If no `past_key_values` are passed, [DynamicCache](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.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. - **pixel_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) -- The tensors corresponding to the input images. Pixel values can be obtained using [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor). See `SmolVLMImageProcessor.__call__()` for details ([SmolVLMProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMProcessor) uses [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor) for processing images). - **pixel_attention_mask** (`torch.Tensor` of shape `(batch_size, image_size, image_size)`, *optional*) -- Mask to avoid performing attention on padding pixel indices. - **image_hidden_states** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`) -- The hidden states of the image encoder after modality projection. - **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`). - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.0.0rc2/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. - **cache_position** (`torch.LongTensor` of shape `(sequence_length)`, *optional*) -- Indices depicting the position of the input sequence tokens in the sequence. Contrarily to `position_ids`, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.0`transformers.models.smolvlm.modeling_smolvlm.SmolVLMBaseModelOutputWithPast` or `tuple(torch.FloatTensor)`A `transformers.models.smolvlm.modeling_smolvlm.SmolVLMBaseModelOutputWithPast` 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 ([SmolVLMConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMConfig)) 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. 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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). 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. - **image_hidden_states** (`tuple(torch.FloatTensor)`, *optional*) -- Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder Inputs fed to the model can have an arbitrary number of images. To account for this, pixel_values fed to the model have image padding -> (batch_size, max_num_images, 3, max_heights, max_widths) where max_num_images is the maximum number of images among the batch_size samples in the batch. Padding images are not needed beyond padding the pixel_values at the entrance of the model. For efficiency, we only pass through the vision_model's forward the real images by discarding the padding images i.e. pixel_values of size (image_batch_size, 3, height, width) where image_batch_size would be 7 when num_images_per_sample=[1, 3, 1, 2] and max_num_images would be 3. **Parameters:** config ([SmolVLMConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMConfig)) : 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()](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** ``transformers.models.smolvlm.modeling_smolvlm.SmolVLMBaseModelOutputWithPast` or `tuple(torch.FloatTensor)`` A `transformers.models.smolvlm.modeling_smolvlm.SmolVLMBaseModelOutputWithPast` 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 ([SmolVLMConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMConfig)) 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. 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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). 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. - **image_hidden_states** (`tuple(torch.FloatTensor)`, *optional*) -- Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder ## SmolVLMForConditionalGeneration[[transformers.SmolVLMForConditionalGeneration]][[transformers.SmolVLMForConditionalGeneration]] #### transformers.SmolVLMForConditionalGeneration[[transformers.SmolVLMForConditionalGeneration]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/modeling_smolvlm.py#L709) The SmolVLM Model with a language modeling head. It is made up a SigLIP vision encoder, with a language modeling head on top. This model inherits from [PreTrainedModel](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.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](https://pytorch.org/docs/stable/nn.html#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. forwardtransformers.SmolVLMForConditionalGeneration.forwardhttps://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/modeling_smolvlm.py#L734[{"name": "input_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "position_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "past_key_values", "val": ": typing.Optional[transformers.cache_utils.Cache] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "pixel_values", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "pixel_attention_mask", "val": ": typing.Optional[torch.BoolTensor] = None"}, {"name": "image_hidden_states", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "labels", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "use_cache", "val": ": typing.Optional[bool] = None"}, {"name": "output_attentions", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "cache_position", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}, {"name": "logits_to_keep", "val": ": typing.Union[int, torch.Tensor] = 0"}, {"name": "**kwargs", "val": ": typing_extensions.Unpack[transformers.utils.generic.TransformersKwargs]"}]- **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](/docs/transformers/v5.0.0rc2/ko/model_doc/auto#transformers.AutoTokenizer). See [PreTrainedTokenizer.encode()](/docs/transformers/v5.0.0rc2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.encode) and [PreTrainedTokenizer.__call__()](/docs/transformers/v5.0.0rc2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase.__call__) for details. [What are input IDs?](../glossary#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?](../glossary#attention-mask) - **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?](../glossary#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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance is allowed as input, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). If no `past_key_values` are passed, [DynamicCache](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.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. - **pixel_values** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`, *optional*) -- The tensors corresponding to the input images. Pixel values can be obtained using [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor). See `SmolVLMImageProcessor.__call__()` for details ([SmolVLMProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMProcessor) uses [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor) for processing images). - **pixel_attention_mask** (`torch.Tensor` of shape `(batch_size, image_size, image_size)`, *optional*) -- Mask to avoid performing attention on padding pixel indices. - **image_hidden_states** (`torch.FloatTensor` of shape `(batch_size, num_channels, image_size, image_size)`) -- The hidden states of the image encoder after modality projection. - **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 `model.image_token_id`. Tokens with indices set to `model.image_token_id` are ignored (masked), the loss is only computed for the 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`). - **output_attentions** (`bool`, *optional*) -- Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned tensors for more detail. - **output_hidden_states** (`bool`, *optional*) -- Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for more detail. - **cache_position** (`torch.LongTensor` of shape `(sequence_length)`, *optional*) -- Indices depicting the position of the input sequence tokens in the sequence. Contrarily to `position_ids`, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length. - **return_dict** (`bool`, *optional*) -- Whether or not to return a [ModelOutput](/docs/transformers/v5.0.0rc2/ko/main_classes/output#transformers.utils.ModelOutput) instead of a plain tuple. - **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).0`transformers.models.smolvlm.modeling_smolvlm.SmolVLMCausalLMOutputWithPast` or `tuple(torch.FloatTensor)`A `transformers.models.smolvlm.modeling_smolvlm.SmolVLMCausalLMOutputWithPast` 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 ([SmolVLMConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMConfig)) and inputs. - **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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). 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. - **image_hidden_states** (`tuple(torch.FloatTensor)`, *optional*) -- Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder The [SmolVLMForConditionalGeneration](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration) 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. Example: ```python >>> import requests >>> import torch >>> from PIL import Image >>> from io import BytesIO >>> from transformers import AutoProcessor, AutoModelForImageTextToText >>> from transformers.image_utils import load_image >>> # Note that passing the image urls (instead of the actual pil images) to the processor is also possible >>> image1 = load_image("https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg") >>> image2 = load_image("https://cdn.britannica.com/59/94459-050-DBA42467/Skyline-Chicago.jpg") >>> image3 = load_image("https://cdn.britannica.com/68/170868-050-8DDE8263/Golden-Gate-Bridge-San-Francisco.jpg") >>> processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM2-2.2B-Instruct") >>> model = AutoModelForImageTextToText.from_pretrained("HuggingFaceTB/SmolVLM2-2.2B-Instruct", dtype=torch.bfloat16, device_map="auto") >>> # Create inputs >>> messages = [ ... { ... "role": "user", ... "content": [ ... {"type": "video", "path": path/to/video}, ... {"type": "text", "text": "What is happening in this video?"}, ... ] ... } ... ] >>> inputs = processor.apply_chat_template([messages], add_generation_prompt=True) >>> # Generate >>> generated_ids = model.generate(**inputs, max_new_tokens=256) >>> generated_texts = processor.batch_decode(generated_ids, skip_special_tokens=True) >>> print(generated_texts) ``` **Parameters:** config ([SmolVLMForConditionalGeneration](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMForConditionalGeneration)) : 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()](/docs/transformers/v5.0.0rc2/ko/main_classes/model#transformers.PreTrainedModel.from_pretrained) method to load the model weights. **Returns:** ``transformers.models.smolvlm.modeling_smolvlm.SmolVLMCausalLMOutputWithPast` or `tuple(torch.FloatTensor)`` A `transformers.models.smolvlm.modeling_smolvlm.SmolVLMCausalLMOutputWithPast` 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 ([SmolVLMConfig](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMConfig)) and inputs. - **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](/docs/transformers/v5.0.0rc2/ko/internal/generation_utils#transformers.Cache) instance. For more details, see our [kv cache guide](https://huggingface.co/docs/transformers/en/kv_cache). 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. - **image_hidden_states** (`tuple(torch.FloatTensor)`, *optional*) -- Tuple of `torch.FloatTensor` (one for the output of the image embeddings, `(batch_size, num_images, sequence_length, hidden_size)`. image_hidden_states of the model produced by the vision encoder ## SmolVLMImageProcessor[[transformers.SmolVLMImageProcessor]][[transformers.SmolVLMImageProcessor]] #### transformers.SmolVLMImageProcessor[[transformers.SmolVLMImageProcessor]] [Source](https://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/image_processing_smolvlm.py#L268) Constructs a SmolVLM image processor. preprocesstransformers.SmolVLMImageProcessor.preprocesshttps://github.com/huggingface/transformers/blob/v5.0.0rc2/src/transformers/models/smolvlm/image_processing_smolvlm.py#L620[{"name": "images", "val": ": typing.Union[ForwardRef('PIL.Image.Image'), numpy.ndarray, ForwardRef('torch.Tensor'), list['PIL.Image.Image'], list[numpy.ndarray], list['torch.Tensor']]"}, {"name": "do_convert_rgb", "val": ": typing.Optional[bool] = None"}, {"name": "do_resize", "val": ": typing.Optional[bool] = None"}, {"name": "size", "val": ": typing.Optional[dict[str, int]] = None"}, {"name": "resample", "val": ": typing.Optional[PIL.Image.Resampling] = None"}, {"name": "do_image_splitting", "val": ": typing.Optional[bool] = None"}, {"name": "do_rescale", "val": ": typing.Optional[bool] = None"}, {"name": "max_image_size", "val": ": typing.Optional[dict[str, int]] = None"}, {"name": "rescale_factor", "val": ": typing.Optional[float] = None"}, {"name": "do_normalize", "val": ": typing.Optional[bool] = None"}, {"name": "image_mean", "val": ": typing.Union[float, list[float], NoneType] = None"}, {"name": "image_std", "val": ": typing.Union[float, list[float], NoneType] = None"}, {"name": "do_pad", "val": ": typing.Optional[bool] = None"}, {"name": "return_tensors", "val": ": typing.Union[str, transformers.utils.generic.TensorType, NoneType] = None"}, {"name": "return_row_col_info", "val": ": bool = False"}, {"name": "data_format", "val": ": typing.Optional[transformers.image_utils.ChannelDimension] = "}, {"name": "input_data_format", "val": ": typing.Union[str, transformers.image_utils.ChannelDimension, NoneType] = None"}]- **images** (`ImageInput`) -- A list of images to preprocess. - **do_convert_rgb** (`bool`, *optional*, defaults to `self.do_convert_rgb`) -- Whether to convert the image to RGB. - **do_resize** (`bool`, *optional*, defaults to `self.do_resize`) -- Whether to resize the image. - **size** (`dict[str, int]`, *optional*, defaults to `self.size`) -- Size of the image after resizing. With the longest edge resized to keep the input aspect ratio. - **resample** (`int`, *optional*, defaults to `self.resample`) -- Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only has an effect if `do_resize` is set to `True`. - **do_image_splitting** (`bool`, *optional*, defaults to `self.do_image_splitting`) -- Whether to split the image into sub-images concatenated with the original image. They are split into patches such that each patch has a size of `max_image_size["height"]` x `max_image_size["width"]`. - **max_image_size** (`Dict`, *optional*, defaults to `self.max_image_size`) -- Maximum resolution of the images. If the image is larger than this size, the image is split into patches. - **do_rescale** (`bool`, *optional*, defaults to `self.do_rescale`) -- Whether to rescale the image. - **rescale_factor** (`float`, *optional*, defaults to `self.rescale_factor`) -- Rescale factor to rescale the image by if `do_rescale` is set to `True`. - **do_normalize** (`bool`, *optional*, defaults to `self.do_normalize`) -- Whether to normalize the image. - **image_mean** (`float` or `list[float]`, *optional*, defaults to `self.image_mean`) -- Image mean to use for normalization. Only has an effect if `do_normalize` is set to `True`. - **image_std** (`float` or `list[float]`, *optional*, defaults to `self.image_std`) -- Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to `True`. - **do_pad** (`bool`, *optional*, defaults to `self.do_pad`) -- Whether or not to pad the images to the largest height and width in the batch. - **return_tensors** (`str` or `TensorType`, *optional*) -- The type of tensors to return. Can be one of: - Unset: Return a list of `np.ndarray`. - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`. - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`. - **return_row_col_info** (`bool`, *optional*, default to `False`) -- Whether to return the number of rows and columns of the split images. This is used for the `SmolVLMProcessor` to generate prompt strings based on the number of rows and columns. - **data_format** (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`) -- The channel dimension format for the output image. Can be one of: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - Unset: Use the channel dimension format of the input image. - **input_data_format** (`ChannelDimension` or `str`, *optional*) -- The channel dimension format for the input image. If unset, the channel dimension format is inferred from the input image. Can be one of: - `"channels_first"` or `ChannelDimension.FIRST`: image in (num_channels, height, width) format. - `"channels_last"` or `ChannelDimension.LAST`: image in (height, width, num_channels) format. - `"none"` or `ChannelDimension.NONE`: image in (height, width) format.0 Preprocess a batch of images. **Parameters:** do_convert_rgb (`bool`, *optional*, defaults to `True`) : Whether to convert the image to RGB. This is useful if the input image is of a different format e.g. RGBA. Only has an effect if the input image is in the PIL format. do_resize (`bool`, *optional*, defaults to `True`) : Whether to resize the image. The longest edge of the image is resized to be >> import requests >>> from transformers import SmolVLMProcessor >>> from transformers.image_utils import load_image >>> processor = SmolVLMProcessor.from_pretrained("HuggingFaceM4/SmolVLM2-256M-Video-Instruct") >>> processor.image_processor.do_image_splitting = False # Force as False to simplify the example >>> url1 = "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" >>> url2 = "https://cdn.britannica.com/59/94459-050-DBA42467/Skyline-Chicago.jpg" >>> image1, image2 = load_image(url1), load_image(url2) >>> images = [[image1], [image2]] >>> text = [ ... "In this image, we see", ... "bla bla bla", ... ] >>> outputs = processor(images=images, text=text, return_tensors="pt", padding=True) >>> input_ids = outputs.input_ids >>> input_tokens = processor.tokenizer.batch_decode(input_ids) >>> print(input_tokens) ['(()*169) In this image, we see', 'bla bla bla(()*169)'] ``` **Parameters:** image_processor (`SmolVLMImageProcessor`) : An instance of [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor). The image processor is a required input. tokenizer (`PreTrainedTokenizerBase`) : An instance of [PreTrainedTokenizerBase](/docs/transformers/v5.0.0rc2/ko/internal/tokenization_utils#transformers.PreTrainedTokenizerBase). This should correspond with the model's text model. The tokenizer is a required input. video_processor (`SmolVLMImageProcessor`) : n instance of [SmolVLMImageProcessor](/docs/transformers/v5.0.0rc2/ko/model_doc/smolvlm#transformers.SmolVLMImageProcessor). The video processor is a required input. image_seq_len (`int`, *optional*, defaults to 169) : The length of the image sequence i.e. the number of tokens per image in the input. This parameter is used to build the string from the input prompt and image tokens and should match the value the model used. It is computed as: image_seq_len = int(((image_size // patch_size) ** 2) / (scale_factor**2)) chat_template (`str`, *optional*) : A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string.