Transformers documentation

Qwen2-VL

Transformers

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

Qwen2-VL

Overview

The Qwen2-VL model is a major update to Qwen-VL from the Qwen team at Alibaba Research.

The abstract from the blog is the following:

This blog introduces Qwen2-VL, an advanced version of the Qwen-VL model that has undergone significant enhancements over the past year. Key improvements include enhanced image comprehension, advanced video understanding, integrated visual agent functionality, and expanded multilingual support. The model architecture has been optimized for handling arbitrary image resolutions through Naive Dynamic Resolution support and utilizes Multimodal Rotary Position Embedding (M-ROPE) to effectively process both 1D textual and multi-dimensional visual data. This updated model demonstrates competitive performance against leading AI systems like GPT-4o and Claude 3.5 Sonnet in vision-related tasks and ranks highly among open-source models in text capabilities. These advancements make Qwen2-VL a versatile tool for various applications requiring robust multimodal processing and reasoning abilities.

Qwen2-VL architecture. Taken from the blog post.

This model was contributed by simonJJJ.

Usage example

Single Media inference

The model can accept both images and videos as input. Here’s an example code for inference.


from PIL import Image
import requests
import torch
from torchvision import io
from typing import Dict
from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor

# Load the model in half-precision on the available device(s)
model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", device_map="auto")
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")

# Image
url = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"
image = Image.open(requests.get(url, stream=True).raw)

conversation = [
    {
        "role":"user",
        "content":[
            {
                "type":"image",
            },
            {
                "type":"text",
                "text":"Describe this image."
            }
        ]
    }
]


# Preprocess the inputs
text_prompt = processor.apply_chat_template(conversation, add_generation_prompt=True)
# Excepted output: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>\n<|im_start|>assistant\n'

inputs = processor(text=[text_prompt], images=[image], padding=True, return_tensors="pt")
inputs = inputs.to('cuda')

# Inference: Generation of the output
output_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, output_ids)]
output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
print(output_text)

# Video
def fetch_video(ele: Dict, nframe_factor=2):
    if isinstance(ele['video'], str):
        def round_by_factor(number: int, factor: int) -> int:
            return round(number / factor) * factor

        video = ele["video"]
        if video.startswith("file://"):
            video = video[7:]

        video, _, info = io.read_video(
            video,
            start_pts=ele.get("video_start", 0.0),
            end_pts=ele.get("video_end", None),
            pts_unit="sec",
            output_format="TCHW",
        )
        assert not ("fps" in ele and "nframes" in ele), "Only accept either `fps` or `nframes`"
        if "nframes" in ele:
            nframes = round_by_factor(ele["nframes"], nframe_factor)
        else:
            fps = ele.get("fps", 1.0)
            nframes = round_by_factor(video.size(0) / info["video_fps"] * fps, nframe_factor)
        idx = torch.linspace(0, video.size(0) - 1, nframes, dtype=torch.int64)
        return video[idx]

video_info = {"type": "video", "video": "/path/to/video.mp4", "fps": 1.0}
video = fetch_video(video_info)
conversation = [
    {
        "role": "user",
        "content": [
            {"type": "video"},
            {"type": "text", "text": "What happened in the video?"},
        ],
    }
]

# Preprocess the inputs
text_prompt = processor.apply_chat_template(conversation, add_generation_prompt=True)
# Excepted output: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|video_pad|><|vision_end|>What happened in the video?<|im_end|>\n<|im_start|>assistant\n'

inputs = processor(text=[text_prompt], videos=[video], padding=True, return_tensors="pt")
inputs = inputs.to('cuda')

# Inference: Generation of the output
output_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, output_ids)]
output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
print(output_text)

Batch Mixed Media Inference

The model can batch inputs composed of mixed samples of various types such as images, videos, and text. Here is an example.

image1 = Image.open("/path/to/image1.jpg")
image2 = Image.open("/path/to/image2.jpg")
image3 = Image.open("/path/to/image3.jpg")
image4 = Image.open("/path/to/image4.jpg")
image5 = Image.open("/path/to/image5.jpg")
video = fetch_video({
    "type": "video",
    "video": "/path/to/video.mp4",
    "fps": 1.0
})

# Conversation for the first image
conversation1 = [
    {
        "role": "user",
        "content": [
            {"type": "image"},
            {"type": "text", "text": "Describe this image."}
        ]
    }
]

# Conversation with two images
conversation2 = [
    {
        "role": "user",
        "content": [
            {"type": "image"},
            {"type": "image"},
            {"type": "text", "text": "What is written in the pictures?"}
        ]
    }
]

# Conversation with pure text
conversation3 = [
    {
        "role": "user",
        "content": "who are you?"
    }
]


# Conversation with mixed midia
conversation4 = [
    {
        "role": "user",
        "content": [
            {"type": "image"},
            {"type": "image"},
            {"type": "video"},
            {"type": "text", "text": "What are the common elements in these medias?"},
        ],
    }
]

conversations = [conversation1, conversation2, conversation3, conversation4]
# Preparation for batch inference
texts = [processor.apply_chat_template(msg, add_generation_prompt=True) for msg in conversations]
inputs = processor(
    text=texts,
    images=[image1, image2, image3, image4, image5],
    videos=[video],
    padding=True,
    return_tensors="pt",
)
inputs = inputs.to('cuda')

# Batch Inference
output_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, output_ids)]
output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
print(output_text)

Usage Tips

Image Resolution trade-off

The model supports a wide range of resolution inputs. By default, it uses the native resolution for input, but higher resolutions can enhance performance at the cost of more computation. Users can set the minimum and maximum number of pixels to achieve an optimal configuration for their needs.

min_pixels = 224*224
max_pixels = 2048*2048
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)

In case of limited GPU RAM, one can reduce the resolution as follows:

min_pixels = 256*28*28
max_pixels = 1024*28*28 
processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)

This ensures each image gets encoded using a number between 256-1024 tokens. The 28 comes from the fact that the model uses a patch size of 14 and a temporal patch size of 2 (14 x 2 = 28).

Multiple Image Inputs

By default, images and video content are directly included in the conversation. When handling multiple images, it’s helpful to add labels to the images and videos for better reference. Users can control this behavior with the following settings:

conversation = [
    {
        "role": "user",
        "content": [
            {"type": "image"}, 
            {"type": "text", "text": "Hello, how are you?"}
        ]
    },
    {
        "role": "assistant",
        "content": "I'm doing well, thank you for asking. How can I assist you today?"
    },
    {
        "role": "user",
        "content": [
            {"type": "text", "text": "Can you describe these images and video?"}, 
            {"type": "image"}, 
            {"type": "image"}, 
            {"type": "video"}, 
            {"type": "text", "text": "These are from my vacation."}
        ]
    },
    {
        "role": "assistant",
        "content": "I'd be happy to describe the images and video for you. Could you please provide more context about your vacation?"
    },
    {
        "role": "user",
        "content": "It was a trip to the mountains. Can you see the details in the images and video?"
    }
]

# default:
prompt_without_id = processor.apply_chat_template(conversation, add_generation_prompt=True)
# Excepted output: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Hello, how are you?<|im_end|>\n<|im_start|>assistant\nI'm doing well, thank you for asking. How can I assist you today?<|im_end|>\n<|im_start|>user\nCan you describe these images and video?<|vision_start|><|image_pad|><|vision_end|><|vision_start|><|image_pad|><|vision_end|><|vision_start|><|video_pad|><|vision_end|>These are from my vacation.<|im_end|>\n<|im_start|>assistant\nI'd be happy to describe the images and video for you. Could you please provide more context about your vacation?<|im_end|>\n<|im_start|>user\nIt was a trip to the mountains. Can you see the details in the images and video?<|im_end|>\n<|im_start|>assistant\n'


# add ids
prompt_with_id = processor.apply_chat_template(conversation, add_generation_prompt=True, add_vision_id=True)
# Excepted output: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nPicture 1: <|vision_start|><|image_pad|><|vision_end|>Hello, how are you?<|im_end|>\n<|im_start|>assistant\nI'm doing well, thank you for asking. How can I assist you today?<|im_end|>\n<|im_start|>user\nCan you describe these images and video?Picture 2: <|vision_start|><|image_pad|><|vision_end|>Picture 3: <|vision_start|><|image_pad|><|vision_end|>Video 1: <|vision_start|><|video_pad|><|vision_end|>These are from my vacation.<|im_end|>\n<|im_start|>assistant\nI'd be happy to describe the images and video for you. Could you please provide more context about your vacation?<|im_end|>\n<|im_start|>user\nIt was a trip to the mountains. Can you see the details in the images and video?<|im_end|>\n<|im_start|>assistant\n'

Flash-Attention 2 to speed up generation

First, make sure to install the latest version of Flash Attention 2:

pip install -U flash-attn --no-build-isolation

Also, you should have a hardware that is compatible with Flash-Attention 2. Read more about it in the official documentation of the flash attention repository. FlashAttention-2 can only be used when a model is loaded in torch.float16 or torch.bfloat16.

To load and run a model using Flash Attention-2, simply add attn_implementation="flash_attention_2" when loading the model as follows:

from transformers import Qwen2VLForConditionalGeneration

model = Qwen2VLForConditionalGeneration.from_pretrained(
    "Qwen/Qwen2-VL-7B-Instruct", 
    torch_dtype=torch.bfloat16, 
    attn_implementation="flash_attention_2",
)

Qwen2VLConfig

class transformers.Qwen2VLConfig

< source >

( vocab_size = 152064 hidden_size = 8192 intermediate_size = 29568 num_hidden_layers = 80 num_attention_heads = 64 num_key_value_heads = 8 hidden_act = 'silu' max_position_embeddings = 32768 initializer_range = 0.02 rms_norm_eps = 1e-05 use_cache = True tie_word_embeddings = False rope_theta = 1000000.0 use_sliding_window = False sliding_window = 4096 max_window_layers = 80 attention_dropout = 0.0 vision_config = None rope_scaling = None **kwargs )

Parameters

vocab_size (int, optional, defaults to 152064) — Vocabulary size of the Qwen2VL model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling Qwen2VLModel
hidden_size (int, optional, defaults to 8192) — Dimension of the hidden representations.
intermediate_size (int, optional, defaults to 29568) — Dimension of the MLP representations.
num_hidden_layers (int, optional, defaults to 80) — Number of hidden layers in the Transformer encoder.
num_attention_heads (int, optional, defaults to 64) — Number of attention heads for each attention layer in the Transformer encoder.
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 checkout this paper. If it is not specified, will default to 32.
hidden_act (str or function, optional, defaults to "silu") — The non-linear activation function (function or string) in the decoder.
max_position_embeddings (int, optional, defaults to 32768) — 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.
tie_word_embeddings (bool, optional, defaults to False) — Whether the model’s input and output word embeddings should be tied.
rope_theta (float, optional, defaults to 1000000.0) — The base period of the RoPE embeddings.
use_sliding_window (bool, optional, defaults to False) — Whether to use sliding window attention.
sliding_window (int, optional, defaults to 4096) — Sliding window attention (SWA) window size. If not specified, will default to 4096.
max_window_layers (int, optional, defaults to 80) — The number of layers that use SWA (Sliding Window Attention). The bottom layers use SWA while the top use full attention.
attention_dropout (float, optional, defaults to 0.0) — The dropout ratio for the attention probabilities.
vision_config (Dict, optional) — The config for the visual encoder initialization.
rope_scaling (Dict, optional) — Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer max_position_embeddings, we recommend you to update this value accordingly. Expected contents: rope_type (str): The sub-variant of RoPE to use. Can be one of [‘default’, ‘linear’, ‘dynamic’, ‘yarn’, ‘longrope’, ‘llama3’], with ‘default’ being the original RoPE implementation. factor (float, optional): Used with all rope types except ‘default’. The scaling factor to apply to the RoPE embeddings. In most scaling types, a factor of x will enable the model to handle sequences of length x original maximum pre-trained length. original_max_position_embeddings (int, optional): Used with ‘dynamic’, ‘longrope’ and ‘llama3’. The original max position embeddings used during pretraining. attention_factor (float, optional): Used with ‘yarn’ and ‘longrope’. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the factor field to infer the suggested value. beta_fast (float, optional): Only used with ‘yarn’. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. beta_slow (float, optional): Only used with ‘yarn’. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. short_factor (List[float], optional): Only used with ‘longrope’. The scaling factor to be applied to short contexts (< original_max_position_embeddings). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 long_factor (List[float], optional): Only used with ‘longrope’. The scaling factor to be applied to long contexts (< original_max_position_embeddings). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 low_freq_factor (float, optional): Only used with ‘llama3’. Scaling factor applied to low frequency components of the RoPE high_freq_factor (float, optional*): Only used with ‘llama3’. Scaling factor applied to high frequency components of the RoPE

This is the configuration class to store the configuration of a Qwen2VLModel. It is used to instantiate a Qwen2-VL model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of Qwen2-VL-7B-Instruct Qwen/Qwen2-VL-7B-Instruct.

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 Qwen2VLForConditionalGeneration, Qwen2VLConfig

>>> # Initializing a Qwen2VL style configuration
>>> configuration = Qwen2VLConfig()

>>> # Initializing a model from the Qwen2-VL-7B style configuration
>>> model = Qwen2VLForConditionalGeneration(configuration)

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

Qwen2VLImageProcessor

class transformers.Qwen2VLImageProcessor

< source >

( do_resize: bool = True resample: Resampling = <Resampling.BICUBIC: 3> do_rescale: bool = True rescale_factor: Union = 0.00392156862745098 do_normalize: bool = True image_mean: Union = None image_std: Union = None do_convert_rgb: bool = True min_pixels: int = 3136 max_pixels: int = 1003520 patch_size: int = 14 temporal_patch_size: int = 2 merge_size: int = 2 **kwargs )

Parameters

do_resize (bool, optional, defaults to True) — Whether to resize the image’s (height, width) dimensions.
resample (PILImageResampling, optional, defaults to Resampling.BICUBIC) — Resampling filter to use when resizing the image.
do_rescale (bool, optional, defaults to True) — Whether to rescale the image by the specified scale rescale_factor.
rescale_factor (int or float, optional, defaults to 1/255) — Scale factor to use if rescaling the image.
do_normalize (bool, optional, defaults to True) — Whether to normalize the image.
image_mean (float or List[float], optional, defaults to [0.48145466, 0.4578275, 0.40821073]) — Mean to use if normalizing the image. This is a float or list of floats for each channel in the image.
image_std (float or List[float], optional, defaults to [0.26862954, 0.26130258, 0.27577711]) — Standard deviation to use if normalizing the image. This is a float or list of floats for each channel in the image.
do_convert_rgb (bool, optional, defaults to True) — Whether to convert the image to RGB.
min_pixels (int, optional, defaults to 56 * 56) — The min pixels of the image to resize the image.
max_pixels (int, optional, defaults to 28 * 28 * 1280) — The max pixels of the image to resize the image.
patch_size (int, optional, defaults to 14) — The spacial patch size of the vision encoder.
temporal_patch_size (int, optional, defaults to 2) — The temporal patch size of the vision encoder.
merge_size (int, optional, defaults to 2) — The merge size of the vision encoder to llm encoder.

Constructs a Qwen2-VL image processor that dynamically resizes images based on the original images.

preprocess

< source >

( images: Union videos: Union = None do_resize: bool = None size: Dict = None resample: Resampling = None do_rescale: bool = None rescale_factor: float = None do_normalize: bool = None image_mean: Union = None image_std: Union = None do_convert_rgb: bool = None return_tensors: Union = None data_format: Optional = <ChannelDimension.FIRST: 'channels_first'> input_data_format: Union = None )

Parameters

images (ImageInput) — 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.
videos (VideoInput) — 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.
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. Shortest edge of the image is resized to size[“shortest_edge”], 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_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_convert_rgb (bool, optional, defaults to self.do_convert_rgb) — Whether to convert the image to RGB.
return_tensors (str or TensorType, optional) — The type of tensors to return. Can be one of:
- Unset: Return a list of np.ndarray.
- TensorType.TENSORFLOW or 'tf': Return a batch of type tf.Tensor.
- TensorType.PYTORCH or 'pt': Return a batch of type torch.Tensor.
- TensorType.NUMPY or 'np': Return a batch of type np.ndarray.
- TensorType.JAX or 'jax': Return a batch of type jax.numpy.ndarray.
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.

Qwen2VLProcessor

class transformers.Qwen2VLProcessor

< source >

( image_processor = None tokenizer = None chat_template = None **kwargs )

Parameters

image_processor (Qwen2VLImageProcessor, optional) — The image processor is a required input.
tokenizer (Qwen2TokenizerFast, optional) — The tokenizer is a required input.
chat_template (str, optional) — A Jinja template which will be used to convert lists of messages in a chat into a tokenizable string.

Constructs a Qwen2-VL processor which wraps a Qwen2-VL image processor and a Qwen2 tokenizer into a single processor. Qwen2VLProcessor offers all the functionalities of Qwen2VLImageProcessor and Qwen2TokenizerFast. See the __call__() and decode() for more information.

batch_decode

< source >

( *args **kwargs )

This method forwards all its arguments to Qwen2TokenizerFast’s batch_decode(). Please refer to the docstring of this method for more information.

decode

< source >

( *args **kwargs )

This method forwards all its arguments to Qwen2TokenizerFast’s decode(). Please refer to the docstring of this method for more information.

Qwen2VLModel

class transformers.Qwen2VLModel

< source >

( config: Qwen2VLConfig )

Parameters

config (Qwen2VLConfig) — 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 Qwen2VL 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

< source >

( input_ids: LongTensor = None attention_mask: Optional = None position_ids: Optional = None past_key_values: Optional = None inputs_embeds: Optional = None use_cache: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None cache_position: Optional = None )

Qwen2VLForConditionalGeneration

class transformers.Qwen2VLForConditionalGeneration

< source >

( config )

forward

< source >

( input_ids: LongTensor = None attention_mask: Optional = None position_ids: Optional = None past_key_values: Optional = None inputs_embeds: Optional = None labels: Optional = None use_cache: Optional = None output_attentions: Optional = None output_hidden_states: Optional = None return_dict: Optional = None pixel_values: Optional = None pixel_values_videos: Optional = None image_grid_thw: Optional = None video_grid_thw: Optional = None rope_deltas: Optional = None ) → transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLCausalLMOutputWithPast or tuple(torch.FloatTensor)

Parameters

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

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?

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

If past_key_values is used, optionally only the last decoder_input_ids have to be input (see past_key_values).

If you want to change padding behavior, you should read modeling_opt._prepare_decoder_attention_mask and modify to your needs. See diagram 1 in the paper for more information on the default strategy.
- 1 indicates the head is not masked,
- 0 indicates the head is masked.
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 (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head)) and 2 additional tensors of shape (batch_size, num_heads, encoder_sequence_length, embed_size_per_head).

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

If past_key_values are used, the user can optionally input only the last decoder_input_ids (those that don’t have their past key value states given to this model) of shape (batch_size, 1) instead of all decoder_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).
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 instead of a plain tuple.
pixel_values (torch.FloatTensor of shape `(seq_length, num_channels image_size image_size)) — The tensors corresponding to the input images. Pixel values can be obtained using AutoImageProcessor. See Qwen2VLImageProcessor.call() for details. Qwen2VLProcessor uses Qwen2VLImageProcessor for processing images.
pixel_values_videos (torch.FloatTensor of shape `(seq_length, num_channels temporal_size image_size * image_size)) — The tensors corresponding to the input videos. Pixel values can be obtained using AutoImageProcessor. See Qwen2VLImageProcessor.call() for details. Qwen2VLProcessor uses Qwen2VLImageProcessor for processing videos.
image_grid_thw (torch.LongTensor of shape (num_images, 3), optional) — The temporal, height and width of feature shape of each image in LLM.
video_grid_thw (torch.LongTensor of shape (num_videos, 3), optional) — The temporal, height and width of feature shape of each video in LLM.
rope_deltas (torch.LongTensor of shape (batch_size, ), optional) — The rope index difference between sequence length and multimodal rope.

Args — 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), the loss is only computed for the tokens with labels in [0, ..., config.vocab_size].

Returns

transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLCausalLMOutputWithPast or tuple(torch.FloatTensor)

A transformers.models.qwen2_vl.modeling_qwen2_vl.Qwen2VLCausalLMOutputWithPast 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 (Qwen2VLConfig) 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 (tuple(tuple(torch.FloatTensor)), optional, returned when use_cache=True is passed or when config.use_cache=True) — Tuple of tuple(torch.FloatTensor) of length config.n_layers, with each tuple having 2 tensors of shape (batch_size, num_heads, sequence_length, embed_size_per_head))

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.
rope_deltas (torch.LongTensor of shape (batch_size, ), optional) — The rope index difference between sequence length and multimodal rope.

The Qwen2VLForConditionalGeneration 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:

>>> from PIL import Image
>>> import requests
>>> from transformers import AutoProcessor, Qwen2VLForConditionalGeneration

>>> model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
>>> processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")

>>> messages = [
    {
        "role": "user",
        "content": [
            {"type": "image"},
            {"type": "text", "text": "What is shown in this image?"},
        ],
    },
]
>>> url = "https://www.ilankelman.org/stopsigns/australia.jpg"
>>> image = Image.open(requests.get(url, stream=True).raw)

>>> text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
>>> inputs = processor(text=[text], images=[image], vision_infos=[vision_infos])

>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"The image shows a street scene with a red stop sign in the foreground. In the background, there is a large red gate with Chinese characters ..."

< > Update on GitHub

←Qwen2Audio Segment Anything→