Qwen/Qwen3-VL-4B-Instruct download history

Qwen/Qwen3-VL-4B-Instruct is a 4.4B-parameter image text to text model by Qwen. In the last 30 days it was downloaded 3,127,953 times (349,816 in the last 7 days), and 33,831,207 times in total.

It ranks #121 on the Hub by monthly downloads and #20 among image text to text models.

It has 539 likes, 8 of them in the last week.

1,035 models build on Qwen3-VL-4B-Instruct: 322 quantized, 481 fine-tuned, 229 adapters and 3 merges. Together with the original they were downloaded 6,269,021 times in the last 30 days. See the Qwen3-VL-4B-Instruct galaxy.

Most downloaded derivatives

Downloads by month

Downloads in the 30 days up to the end of each month, and the change from the month before.

MonthDownloadsChange
Up to 2026-11-183,127,953-27%
October 20264,271,286+29%
September 20263,302,060-16%
August 20263,915,784+2%
July 20263,820,707-4%
June 20263,980,238+29%
May 20263,081,038+26%
April 20262,444,944+20%
March 20262,030,919+56%
February 20261,303,186+103%
January 2026642,784-8%
December 2025701,517-5%

Its busiest 30 days so far ended on 2026-10-30, with 4,337,928 downloads.

Open Qwen/Qwen3-VL-4B-Instruct on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.