gpt-oss-20b needs about 12.4 GB of memory at MXFP4 (a 11.5 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.
gpt-oss-20b keep-30, MXFP4 quantized, 19.7B parameters
gpt-oss-20b keep-30 is a 19.7B parameter model with 30 of 32 experts retained per layer. It is post‑trained with MXFP4 quantization and designed for local inference within 16GB of memory. The model follows the harmony response format and is licensed under Apache 2.0.
Good at
- Reasoning tasks
- Agentic function calling
- Fine‑tuning
Limits
- Requires harmony format
- Does not pass fidelity bar of parent
- No vision support
Best for
- Local inference on 8‑16GB GPUs
- Customizable reasoning effort
- Agentic applications
Who it’s for. Developers and researchers needing an open‑weight GPT‑style model for reasoning and agentic tasks. The model is suitable for experimentation, fine‑tuning, and commercial deployment under Apache 2.0.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| MXFP4 | 11.5 GB | 12.4 GB | Too big | Get file → |
The MXFP4 quantized file is 11.47 GB and requires about 12.4 GB of VRAM at 8 k context. It is suitable for GPUs with 16 GB or more. No other quantizations are available for this model.
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 (11.5 GB file, about 12.4 GB in use).
Which gpt-oss-20b quant should you choose?
MXFP4 (11.5 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 12.4 GB at MXFP4 with an 8K context window. That is the 11.5 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 12.4 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 11.5 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 Joakimpalm-Zen/gpt-oss-20b-keep30-MXFP4-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 gpt-oss-20b keep-30 on 8 GB VRAM?
No. The MXFP4 quantized model needs about 12.4 GB of VRAM at 8 k context, so 8 GB is insufficient.
Does this model support vision?
No. The model has no vision capabilities.
What license does it use?
Apache 2.0, allowing free use and commercial deployment.