Qwen/Qwen2.5-3B-Instruct download history

Qwen/Qwen2.5-3B-Instruct is a 3.1B-parameter text generation model by Qwen. In the last 30 days it was downloaded 1,863,245 times (371,516 in the last 7 days), and 126,106,606 times in total.

It ranks #209 on the Hub by monthly downloads and #31 among text generation models.

It has 642 likes, 3 of them in the last week.

It is a fine-tune of Qwen/Qwen2.5-3B.

4,557 models build on Qwen2.5-3B-Instruct: 899 quantized, 1,693 fine-tuned, 1,883 adapters and 82 merges. Together with the original they were downloaded 4,236,027 times in the last 30 days. See the Qwen2.5-3B-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 2026-11-181,863,245-42%
October 20263,192,563-22%
September 20264,092,961-43%
August 20267,209,581+25%
July 20265,754,431-24%
June 20267,597,019-28%
May 202610,556,636+14%
April 20269,296,032+19%
March 20267,784,088+21%
February 20266,453,490-59%
January 202615,591,458+103%
December 20257,695,225+13%

Its busiest 30 days so far ended on 2026-01-29, with 15,634,051 downloads.

Open Qwen/Qwen2.5-3B-Instruct on Hugging Face

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