ViLaSR

inclusionAI/ViLaSR

Image Text to Text8.3B parameters
Tracked by OpenModelStats since November 15, 2026View on Hugging Face ↗
Downloads 30D
1,795
Total downloads
32K
Gained 7D
—
Gained 30D
—
Likes
18
Rank · tracked models
#27,918
5919(up 5919 places)this week
Parameters
8.3B
Last source update
Aug 11, 2025
  • It moved from #3084 to #2854 among tracked Image Text to Text models this week.

Downloads over time

Daily gain between consecutive observations

Likes over time

Rank among tracked models

Lower is better

Overview

ViLaSR is a image text to text model published by inclusionAI. OpenModelStats has tracked the model since Nov 15, 2026, most recently observing it 4h ago.

Published
Jun 1, 2025
Last updated
Aug 11, 2025
Library
—
License
—
Task
Image Text to Text
Parameters
8,292,166,656
Spaces
—
Derivative models
—
Velocity 7D/day
—
safetensorsqwen2_5_vlimage-text-to-textconversationalendataset:AntResearchNLP/ViLaSR-dataarxiv:2506.09965base_model:Qwen/Qwen2.5-VL-7B-Instructbase_model:finetune:Qwen/Qwen2.5-VL-7B-Instructregion:us

Publisher

inclusionAIView publisher statistics →

Derived from Qwen/Qwen2.5-VL-7B-Instruct (reported by the source; parent not yet tracked).

Similar models

Models similar to ViLaSR
ModelDL 30D
Qwen2.5-VL-7B-InstructQwen5.3M
Qwen2.5-VL-7B-Instruct-AWQQwen711.8K
Qwen2.5-VL-7B-Instruct-abliteratedhuihui-ai309.1K
UI-TARS-1.5-7BByteDance-Seed229.7K
olmOCR-2-7B-1025-FP8allenai186.7K
olmOCR-2-7B-1025allenai81.3K