janhq/Jan-v2-VL-high download history
janhq/Jan-v2-VL-high is an 8.8B-parameter image text to text model by janhq. In the last 30 days it was downloaded 50 times (12 in the last 7 days), and 4,372 times in total.
It ranks #274,353 on the Hub by monthly downloads and #15,184 among image text to text models.
It has 90 likes.
It is a fine-tune of Qwen/Qwen3-VL-8B-Thinking.
13 models build on Jan-v2-VL-high: 11 quantized, 2 fine-tuned, 0 adapters and 0 merges. Together with the original they were downloaded 19,983 times in the last 30 days. See the Jan-v2-VL-high galaxy.
Most downloaded derivatives
- janhq/Jan-v2-VL-high-gguf (quantized), 16.3K downloads in 30 days
- prithivMLmods/Jan-v2-VL-f32-AIO-GGUF (quantized), 1.5K downloads in 30 days
- mradermacher/Jan-v2-VL-high-i1-GGUF (quantized), 912 downloads in 30 days
- mradermacher/Jan-v2-VL-high-GGUF (quantized), 756 downloads in 30 days
- janhq/Jan-v2-VL-high-8bit-mlx (quantized), 201 downloads in 30 days
- janhq/Jan-v2-VL-high-4bit-mlx (quantized), 106 downloads in 30 days
- janhq/Jan-v2-VL-high-bf16-mlx (finetune), 47 downloads in 30 days
- mlx-community/Jan-v2-VL-high-8bit-mlx (quantized), 26 downloads in 30 days
- mlx-community/Jan-v2-VL-high-4bit-mlx (quantized), 18 downloads in 30 days
- mlx-community/Jan-v2-VL-high-bf16-mlx (finetune), 16 downloads in 30 days
- isogen/Jan-v2-VL-high-exl3-4bpw (quantized), 12 downloads in 30 days
- isogen/Jan-v2-VL-high-exl3-8bpw-h8 (quantized), 7 downloads in 30 days
Open janhq/Jan-v2-VL-high on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.