ColQwen3-VL-Embedding-2B-vdr-lora

ColQwen3-VL-Embedding-2B model trained on Visual Document Retrieval query-document screenshot pairs

This is a Multi-Vector Encoder model finetuned from Qwen/Qwen3-VL-Embedding-2B on the llamaindex-vdr-en-train-preprocessed dataset using the sentence-transformers library. It maps inputs to sequences of 128-dimensional token-level vectors and scores them with late interaction (MaxSim), useful for semantic search with late interaction.

Model Details

Model Description

• Model Type: Multi-Vector Encoder
• Base model: Qwen/Qwen3-VL-Embedding-2B <!-- at revision 9f2f7e710d6d81056aa5c0a4f04764fec6bb7bda -->
• Maximum Sequence Length: 262144 tokens
• Output Dimensionality: 128 dimensions
• Similarity Function: maxsim
• Supported Modalities: Text, Image, Video, Message
• Training Dataset:
  • llamaindex-vdr-en-train-preprocessed
• Language: en
• License: apache-2.0

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Multi-Vector Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Multi-Vector Encoders on Hugging Face

Full Model Architecture

MultiVectorEncoder(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}, 'image': {'method': 'forward', 'method_output_name': 'last_hidden_state'}, 'video': {'method': 'forward', 'method_output_name': 'last_hidden_state'}, 'message': {'method': 'forward', 'method_output_name': 'last_hidden_state', 'format': 'structured'}}, 'module_output_name': 'token_embeddings', 'processing_kwargs': {'chat_template': {'add_generation_prompt': True}}, 'unpad_inputs': False, 'architecture': 'Qwen3VLModel'})
  (1): Dense({'in_features': 2048, 'out_features': 128, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity', 'module_input_name': 'token_embeddings', 'module_output_name': 'token_embeddings'})
  (2): MultiVectorMask({'skiplist_words': [], 'keep_only_token_ids': None})
  (3): Normalize({'module_input_name': 'token_embeddings', 'module_output_name': 'token_embeddings'})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import MultiVectorEncoder

# Download from the 🤗 Hub
model = MultiVectorEncoder("tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora")
# Run inference: each input becomes a sequence of per-token vectors (variable length).
queries = [
    'What are the new anthropological perspectives on development as discussed by Quarles Van Ufford and Giri in 2003?',
]
documents = [
    'https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_0.jpg',
    'https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_1.jpg',
    'https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_2.jpg',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings[0].shape, document_embeddings[0].shape)
# (48, 128) (352, 128)

# Get the MaxSim similarity scores
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[35.4085, 33.2571, 23.4033]])

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* -->

Evaluation

Metrics

Multi Vector Information Retrieval

• Dataset: vdr-eval-hard
• Evaluated with <code>MultiVectorInformationRetrievalEvaluator</code>

<!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

llamaindex-vdr-en-train-preprocessed

• Dataset: llamaindex-vdr-en-train-preprocessed at 2323dab
• Size: 10,000 training samples
• Columns: <code>query</code>, <code>image</code>, and <code>negative_0</code>
• Approximate statistics based on the first 100 samples:

  	query	image	negative_0
  type	string	image	image
  modality	text	image	image
  details	<ul><li>min: 27 tokens</li><li>mean: 35.59 tokens</li><li>max: 50 tokens</li></ul>	<ul><li>min: 1000x756 px</li><li>mean: 1437x1628 px</li><li>max: 2044x1869 px</li></ul>	<ul><li>min: 1008x756 px</li><li>mean: 1450x1637 px</li><li>max: 2044x1851 px</li></ul>

• Samples:

  query	image	negative_0
  <code>What are the new anthropological perspectives on development as discussed by Quarles Van Ufford and Giri in 2003?</code>	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_0.jpg" width="200">	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/example_image_0.jpg" width="200">
  <code>What are the three main positions anthropologists have taken in relation to development, as discussed by David Lewis?</code>	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_1.jpg" width="200">	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/example_image_0.jpg" width="200">
  <code>Who are the three sisters known as the Fates in Greek mythology?</code>	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/image_2.jpg" width="200">	<img src="https://huggingface.co/tomaarsen/ColQwen3-VL-Embedding-2B-vdr-lora/resolve/main/assets/example_image_1.jpg" width="200">

• Loss: <code>CachedMultiVectorMultipleNegativesRankingLoss</code> with these parameters:

  {
      "score_metric": "colbert_scores",
      "mini_batch_size": 1,
      "mini_batch_num_tokens": null,
      "score_mini_batch_size": 1,
      "scale": 1.0,
      "size_average": true,
      "gather_across_devices": false
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 64
• num_train_epochs: 1
• learning_rate: 2e-05
• warmup_steps: 0.1
• bf16: True
• save_only_model: True

All Hyperparameters

<details><summary>Click to expand</summary>

• per_device_train_batch_size: 64
• num_train_epochs: 1
• max_steps: -1
• learning_rate: 2e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0.1
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1.0
• label_smoothing_factor: 0.0
• bf16: True
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: None
• trackio_bucket_id: None
• trackio_static_space_id: None
• per_device_eval_batch_size: 8
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: True
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_static_graph: None
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: None
• fsdp_config: None
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}
• max_length: None

</details>

Training Logs

Training Time

• Training: 2.3 hours
• Evaluation: 30.0 minutes
• Total: 2.8 hours

Framework Versions

• Python: 3.11.13
• Sentence Transformers: 5.7.0.dev0
• Transformers: 5.14.1
• PyTorch: 2.11.0+cu128
• Accelerate: 1.5.2
• Datasets: 3.5.0
• Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

CachedMultiVectorMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->