mlx-community/Qwen2.5-3B-Instruct-4bit download history
mlx-community/Qwen2.5-3B-Instruct-4bit is a 3.1B-parameter text generation model by mlx-community. In the last 30 days it was downloaded 14,184 times (3,634 in the last 7 days), and 491,023 times in total.
It ranks #7,073 on the Hub by monthly downloads and #1,574 among text generation models.
It has 2 likes.
It is a fine-tune of Qwen/Qwen2.5-3B.
8 models build on Qwen2.5-3B-Instruct-4bit: 4 quantized, 2 fine-tuned, 2 adapters and 0 merges. Together with the original they were downloaded 14,712 times in the last 30 days. See the Qwen2.5-3B-Instruct-4bit galaxy.
Most downloaded derivatives
- M37labsorg/ade-extract-3b-v1 (adapter), 223 downloads in 30 days
- KarRak007/qwen2.5-3b-git-ai-commit (quantized), 146 downloads in 30 days
- zMANNNNNNNN/cross-model-state-transfer (finetune), 60 downloads in 30 days
- chendren/qwen2.5-3b-cx-lam (adapter), 48 downloads in 30 days
- inglebhavin98/hinglish-smooth-talker-qwen2.5-3b-experimental (finetune), 24 downloads in 30 days
- marcw12/local-whisper-cleanup-3b (quantized), 11 downloads in 30 days
- rohan-bansode/Qwen-2.5-3B-QLORA (quantized), 11 downloads in 30 days
- rlaschuk/pl-tagger-v1 (quantized), 5 downloads in 30 days
Downloads by month
Downloads in the 30 days up to the end of each month, and the change from the month before.
| Month | Downloads | Change |
|---|---|---|
| Up to 2026-11-18 | 14,184 | +15% |
| October 2026 | 12,331 | -14% |
| September 2026 | 14,281 | -26% |
| August 2026 | 19,262 | -75% |
| July 2026 | 77,771 | +278% |
| June 2026 | 20,591 | -3% |
| May 2026 | 21,131 | -10% |
| April 2026 | 23,563 | -26% |
| March 2026 | 31,753 | +122% |
| February 2026 | 14,278 | +50% |
| January 2026 | 9,518 | +118% |
| December 2025 | 4,368 | -17% |
Its busiest 30 days so far ended on 2025-01-31, with 191,274 downloads.
Open mlx-community/Qwen2.5-3B-Instruct-4bit on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.