Skip to content
ModelListedExternal
Download

Charades Stage2

This is the Stage 2 VideoSearch-R1 checkpoint trained for Charades-STA, presented in the paper VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement.

Sourcehuggingface.co/VideoSearchR1/charades-stage2/tree/c78dc49c6b2cffdf9ae36bd01b738f9b9cdce283
KindText modelBase modelQwen/Qwen3-VL-4B-InstructVersionmodelBgrade

Model source

Kind
Text model
File size
4.57 GB
Base model
Qwen/Qwen3-VL-4B-Instruct
Version
vmodel
Updated
Oct 11, 2026
Downloads
68
Source
Hugging Face

Source description

VideoSearch-R1 Charades-STA Stage 2

This is the Stage 2 VideoSearch-R1 checkpoint trained for Charades-STA, presented in the paper VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement.

Usage

Use with the VideoSearch-R1 codebase:

bash scripts/data_construct/download_preextracted_data.bash charades
EVAL_GPUS=0 bash scripts/inference/inference.bash charades --checkpoint VideoSearchR1/charades-stage2

Sources

1 source
Hugging FacePrimaryVersion c78dc49c6b2cffdf9ae36bd01b738f9b9cdce283
huggingface.co/VideoSearchR1/charades-stage2/tree/c78dc49c6b2cffdf9ae36bd01b738f9b9cdce283

Verified Oct 11

Model artifacts

1 artifact

model.safetensors

safetensors · 4.57 GB · SHA-256 327687cee540…ebbe · Hugging Face

Download file

Source excerpts

3 excerpts
ReadmeAug 2huggingface.co/VideoSearchR1/charades-stage2

--- base_model: Qwen/Qwen3-VL-4B-Instruct library_name: transformers license: apache-2.0 pipeline_tag: video-text-to-text tags: - video-retrieval - temporal-grounding - videosearch-r1 --- # VideoSearch-R1 Charades-STA Stage 2 This is the Stage 2 VideoSearch-R1 checkpoint trained for Charades-STA, presented in the paper [VideoSearch-R1: Iterative Video Retrieval and Reasoning via Soft Query Refinement](https://huggingface.co/papers/2607.00446). - **Project Page:** [https://mlvlab.github.io/VideoSearch-R1/](https://mlvlab.github.io/VideoSearch-R1/) - **Repository:** [https://github.com/mlvlab/VideoSearch-R1](https://github.com/mlvlab/VideoSearch-R1) ## Usage Use with the VideoSearch-R1 codebase: ```bash bash scripts/data_construct/download_preextracted_data.bash charades EVAL_GPUS=0 bash scripts/inference/inference.bash charades --checkpoint VideoSearchR1/charades-stage2 ```

NoteOct 11huggingface.co/VideoSearchR1/charades-stage2

Source context: 68 downloads · 0 likes · Pipeline video-text-to-text · Library transformers · Repo VideoSearchR1/charades-stage2

NoteAug 2huggingface.co/VideoSearchR1/charades-stage2

Source context: 39 downloads · 0 likes · Pipeline video-text-to-text · Library transformers · Repo VideoSearchR1/charades-stage2