Gpt Oss 20b needs about 14.7 GB of memory at MXFP4 (a 13.8 GB file) with an 8K context window, so it fits on a 16 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
Open-weight GPT-OSS 20b with vision support, quantized to MXFP4
GPT-OSS 20b is a 20.9B‑parameter open‑weight model from OpenAI, released under Apache‑2.0. It supports vision via an mmproj add‑on and is quantized with MXFP4, enabling inference on GPUs with around 16 GB of memory.
Good at
- Reasoning and chain‑of‑thought
- Agentic tasks like function calling
- Vision input with mmproj
Limits
- Requires harmony prompt format
- Needs mmproj add‑on for images
- Estimated 14.7 GB memory at 8 k context
Best for
- Local inference on 16 GB GPUs
- Agentic applications needing reasoning
- Vision‑enabled chat models
Who it’s for. Developers and researchers who need an open‑weight GPT‑OSS model with vision capability and a permissive license. Suitable for experimentation, fine‑tuning, and commercial deployment on modern GPUs.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| MXFP4 | 13.8 GB | 14.7 GB | Too big | Get file → |
The only available GGUF build is MXFP4, 13.79 GB in size. It requires roughly 14.7 GB of VRAM at 8 k context, so it fits on GPUs with 16 GB or more. No other quantizations are provided.
What GPU do you need for Gpt Oss 20b?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: the best quant that fits is MXFP4 (13.8 GB file, about 14.7 GB in use).
Which Gpt Oss 20b quant should you choose?
MXFP4 (13.8 GB) is the middle option in this repository. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Gpt Oss 20b
How much VRAM does Gpt Oss 20b need?
About 14.7 GB at MXFP4 with an 8K context window. That is the 13.8 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 Gpt Oss 20b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 14.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Gpt Oss 20b GGUF?
It depends on the quant: MXFP4 is 13.8 GB.
What context length does Gpt Oss 20b support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download Gpt Oss 20b GGUF?
The files are on Hugging Face in the SoAIHQ/gpt-oss-20b-VISION-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 Gpt Oss 20b 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 this model on 8 GB VRAM?
No. The MXFP4 build needs about 14.7 GB of VRAM at 8 k context, so 8 GB is insufficient.
Does it support image input?
Yes, but you must load the mmproj add‑on with --mmproj to enable vision.
What license is it released under?
Apache‑2.0, allowing free use, modification, and commercial deployment.