Mage VL needs about 4.4 GB of memory at Q4_K_M (a 2.7 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 6 quants, from 0.0 GB (F16) to 4.7 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.
Mage VL: Compact 4B multimodal model with codec‑native vision
Mage VL is a 4B multimodal foundation model built on Qwen3 architecture. It uses a codec‑native visual encoder and a Qwen3‑4B causal decoder. The model supports image and video understanding with efficient token usage.
Good at
- Efficient video tokenization
- Codec‑native streaming
- Compact 4B size
Limits
- No vision support beyond images/videos
- Requires 8k context for memory estimates
- License is apache‑2.0
Best for
- Video summarization
- Multimodal QA
- Real‑time streaming inference
Who it’s for. Developers needing a lightweight multimodal model for video and image tasks. Suitable for edge deployments where memory is limited.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| F16 | 0.0 GB | 1.7 GB | Fits | Get file → | |
| F16 | 0.1 GB | 1.8 GB | Fits | Get file → | |
| Q8_0 | 0.1 GB | 1.8 GB | Fits | Get file → | |
| Q8_0 | 0.5 GB | 2.2 GB | Fits | Get file → | |
| Q4_K_M | 2.7 GB | 4.4 GB | Fits | Get file → | |
| Q8_0 | 4.7 GB | 6.4 GB | Fits | Get file → |
For 8 GB VRAM, use the Q4_K_M quant (file size 2.72 GB, estimated 4.4 GB memory). For 12 GB VRAM, the Q8_0 quant (file size 4.69 GB, estimated 6.4 GB memory) is suitable. The F16 variants are smaller but require more VRAM at 8k context.
What GPU do you need for Mage VL?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q8_0 (0.5 GB file, about 2.2 GB in use).
- 6 GB VRAM: the best quant that fits is Q4_K_M (2.7 GB file, about 4.4 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (4.7 GB file, about 6.4 GB in use).
Which Mage VL quant should you choose?
Q4_K_M is the usual starting point for Mage VL: it balances file size and quality well. Here it is a 2.7 GB file.
The Q8_0 file is 0.1 GB, about 0.0 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is F16 at 0.0 GB, with the quality trade-offs that come with fewer bits. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Mage VL
How much VRAM does Mage VL need?
About 4.4 GB at Q4_K_M with an 8K context window. That is the 2.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Mage VL on an 8 GB GPU?
Yes, with Q8_0: the 4.7 GB file needs about 6.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Mage VL GGUF?
It depends on the quant: F16 is 0.0 GB, F16 is 0.1 GB, Q8_0 is 0.1 GB, Q8_0 is 0.5 GB, Q4_K_M is 2.7 GB, Q8_0 is 4.7 GB.
What context length does Mage VL support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.
Where can I download Mage VL GGUF?
The files are on Hugging Face in the JohnTdi/Mage-VL-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Mage VL use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Mage VL on 8 GB VRAM?
Yes, the Q4_K_M quant uses about 4.4 GB memory at 8k context, fitting within 8 GB VRAM.
Does Mage VL support text‑only inference?
Yes, the Qwen3‑4B decoder can process text prompts alone without visual input.
What license applies to Mage VL?
Mage VL is released under the apache‑2.0 license.