Qwen2.5-VL-3B-Instruct-abliterated-6bit
mlx-community/Qwen2.5-VL-3B-Instruct-abliterated-6bit
Vision + text · transformers
6-bit Open on Hugging Face
Parameters
3.8B
Weights (est.)
≈ 3.2 GB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-vlm on an Apple silicon Mac:
pip install mlx-vlmpython -m mlx_vlm.generate --model mlx-community/Qwen2.5-VL-3B-Instruct-abliterated-6bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- Qwen/Qwen2.5-VL-3B-Instruct (finetune)
- Vendor
- Qwen
- Task
- Vision + text
- Quantisation
- 6-bit
- Licence
- —
- Downloads
- 70 in the last 30 days · 519 all time · 0 likes
- Published
- 2025-02-16 · updated 2025-02-25 (1y ago)
- Languages
- en
- Tags
- qwen2_5_vl, multimodal, abliterated, uncensored
Other MLX builds of Qwen/Qwen2.5-VL-3B-Instruct
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen2.5-VL-3B-Instruct-abliterated-4bit | 4-bit | ≈ 2.3 GB | 8 GB+ | 122 |
| Qwen2.5-VL-3B-Instruct-3bit | 3-bit | ≈ 2.7 GB | 8 GB+ | 135 |
| Holo1-3B-3bit | 3-bit | ≈ 2.7 GB | 8 GB+ | 34 |
| Qwen2.5-VL-3B-Instruct-4bit | 4-bit | ≈ 3.1 GB | 8 GB+ | 6.5K |
| Nanonets-OCR2-3B-4bit | 4-bit | ≈ 3.1 GB | 8 GB+ | 98 |
| Holo1-3B-4bit | 4-bit | ≈ 3.1 GB | 8 GB+ | 41 |
| Nanonets-OCR-s-4bit | 4-bit | ≈ 3.1 GB | 8 GB+ | 33 |
| Qwen2.5-VL-3B-Instruct-6bit | 6-bit | ≈ 3.2 GB | 8 GB+ | 156 |
| Qwen2.5-VL-3B-Instruct-abliterated-6bit (this one) | 6-bit | ≈ 3.2 GB | 8 GB+ | 70 |
| Nanonets-OCR2-3B-6bit | 6-bit | ≈ 3.8 GB | 8 GB+ | 42 |
| Holo1-3B-6bit | 6-bit | ≈ 3.8 GB | 8 GB+ | 27 |
| Nanonets-OCR-s-6bit | 6-bit | ≈ 3.8 GB | 8 GB+ | 19 |
| Qwen2.5-VL-3B-Instruct-8bit | 8-bit | ≈ 4.1 GB | 8 GB+ | 246 |
| Qwen2.5-VL-3B-Instruct-abliterated-8bit | 8-bit | ≈ 4.1 GB | 8 GB+ | 85 |
| Nanonets-OCR2-3B-8bit | 8-bit | ≈ 4.6 GB | 8 GB+ | 130 |
| Holo1-3B-8bit | 8-bit | ≈ 4.6 GB | 8 GB+ | 30 |
| Nanonets-OCR-s-8bit | 8-bit | ≈ 4.6 GB | 8 GB+ | 21 |
| Qwen2.5-VL-3B-Instruct-bf16 | bf16 | ≈ 7.5 GB | 16 GB+ | 178 |
| Qwen2.5-VL-3B-Instruct-abliterated | — | ≈ 7.5 GB | 16 GB+ | 99 |
| Nanonets-OCR2-3B-bf16 | bf16 | ≈ 7.5 GB | 16 GB+ | 94 |
| Nanonets-OCR-s-bf16 | bf16 | ≈ 7.5 GB | 16 GB+ | 73 |
| Holo1-3B-bf16 | bf16 | ≈ 7.5 GB | 16 GB+ | 25 |