Sa2VA-Qwen3-VL-2B

ByteDance/Sa2VA-Qwen3-VL-2B

Image Text to Text2.7B parameterstransformersLicense: apache-2.0

Tracked by OpenModelStats since October 15, 2026View on Hugging Face ↗

Downloads 30D: 320
Total downloads: 2,155
Gained 7D: —
Gained 30D: —
Likes: 18
Rank · tracked models: #57,361; 410(down 410 places)this week
Parameters: 2.7B
Last source update: Nov 27, 2025

It moved from #5271 to #5313 among tracked Image Text to Text models this week.

7D 30D 90D 1Y All

Total downloads Gained per day Rolling recent downloads

Downloads over time

Cumulative total downloads observed by OpenModelStats

Likes over time

Rank among tracked models

Lower is better

Overview

Sa2VA-Qwen3-VL-2B is a image text to text model published by ByteDance. OpenModelStats has tracked the model since Oct 15, 2026, most recently observing it Oct 28, 2026.

Published: Nov 27, 2025
Last updated: Nov 27, 2025
Library: transformers
License: apache-2.0
Task: Image Text to Text
Parameters: 2,666,774,834
Spaces: —
Derivative models: —
Velocity 7D/day: —

transformerssafetensorssa2va_chatfeature-extractionSa2VAcustom_codeimage-text-to-textconversationalmultilingualarxiv:2501.04001base_model:OpenGVLab/InternVL3-8Bbase_model:merge:OpenGVLab/InternVL3-8B

Publisher

ByteDanceView publisher statistics →

Derived from OpenGVLab/InternVL3-8B (reported by the source; parent not yet tracked).

Similar models

Models similar to Sa2VA-Qwen3-VL-2B
Model	Task	DL 30D	Likes	Params
InternVL3_5-8B-AWQ-4bitcyankiwi	Image Text to Text	219	1	2.7B
Qwen3.5-4B-NVFP4cosmicproc	Image Text to Text	1,417	4	2.6B
LFM2.5-VL-3B-OptiQ-4bitmlx-community	Image Text to Text	795	2	2.7B
GameScaling-2B-S3THUSI-Lab	Image Text to Text	334	3	2.7B
gemma-4-E4B-it-MLX-8bitunsloth	Image Text to Text	1,147	12	2.6B
Qwen3-VL-8B-Instruct-quantized.w4a16SherlockID365	Image Text to Text	73.9K	1	2.7B