4-bit Open on Hugging Face
Parameters
2.1B
Weights (est.)
≈ 1.6 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/InternVL3-2B-4bit --image photo.jpg --prompt "Describe this image"
Details
- Base model
- OpenGVLab/InternViT-300M-448px-V2_5 (merge)
- Vendor
- OpenGVLab
- Task
- Vision + text
- Quantisation
- 4-bit
- Licence
- qwen
- Downloads
- 121 in the last 30 days · 1.2K all time · 0 likes
- Published
- 2025-04-14 · updated 2025-04-14 (1y ago)
- Languages
- multilingual
- Tags
- internvl_chat, feature-extraction, internvl, custom_code
Other MLX builds of OpenGVLab/InternViT-300M-448px-V2_5
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| InternVL3-1B-4bit | 4-bit | ≈ 965 MB | 8 GB+ | 106 |
| InternVL2_5-1B-4bit | 4-bit | ≈ 965 MB | 8 GB+ | 43 |
| InternVL3-2B-4bit (this one) | 4-bit | ≈ 1.6 GB | 8 GB+ | 121 |
| Kimi-VL-A3B-Thinking-4bit | 4-bit | — | — | 951 |
| InternVL3_5-GPT-OSS-20B-A4B-Preview-4bit | 4-bit | — | — | 104 |