GP

gguf MXFP4 gpt oss 20b Derestricted

gpt-ossGGUF by FelladrinUpdated 2025-12-05
Download on Hugging Face
Parameters20.9B
Context128K
Layers24
Architecturegpt-oss
Smallest quant12.1 GB
Quick answer

gguf MXFP4 gpt oss 20b Derestricted needs about 13.0 GB of memory at MXFP4_MOE (a 12.1 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.

gguf MXFP4 gpt oss 20b Derestricted – 20.9B parameter GPT-OSS model

gguf MXFP4 gpt oss 20b Derestricted is a 20.9 billion‑parameter GPT‑OSS model. It was derived by removing refusal behaviors while preserving reasoning performance. The model supports tool usage and is available under an Apache‑2.0 license.

Good at

  • Tool‑based reasoning
  • Large context handling
  • Uncensored responses

Limits

  • No vision support
  • Requires 13 GB VRAM at 8 k context
  • License is Apache‑2.0

Best for

  • Chatbot development
  • Code generation with tools
  • Research on uncensored LLMs

Who it’s for. Developers and researchers needing a large GPT‑OSS model without refusal behavior. The model is suitable for applications that require tool integration and extensive context handling.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
MXFP4_MOE12.1 GB 13.0 GB
Too big Get file →

The only available quant is MXFP4_MOE with a file size of 12.11 GB. It requires approximately 13 GB of VRAM at 8 k context. For lower VRAM budgets, a different quant is not available.

What GPU do you need for gguf MXFP4 gpt oss 20b Derestricted?

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_MOE (12.1 GB file, about 13.0 GB in use).

Which gguf MXFP4 gpt oss 20b Derestricted quant should you choose?

MXFP4_MOE (12.1 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

Architecturegpt-oss
Parameters20.9B (approximate, mixture of experts)
Layers24
Trained context131,072 tokens
KV cache (fp16)48 KB per token

Source

Quantized byFelladrin
Familygpt-oss
Licenseapache-2.0
Downloads (HF)856
Last updated2025-12-05

Frequently asked questions about gguf MXFP4 gpt oss 20b Derestricted

How much VRAM does gguf MXFP4 gpt oss 20b Derestricted need?

About 13.0 GB at MXFP4_MOE with an 8K context window. That is the 12.1 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 gguf MXFP4 gpt oss 20b Derestricted on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 13.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of gguf MXFP4 gpt oss 20b Derestricted GGUF?

It depends on the quant: MXFP4_MOE is 12.1 GB.

What context length does gguf MXFP4 gpt oss 20b Derestricted 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 gguf MXFP4 gpt oss 20b Derestricted GGUF?

The files are on Hugging Face in the Felladrin/gguf-MXFP4-gpt-oss-20b-Derestricted repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does gguf MXFP4 gpt oss 20b Derestricted 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_MOE quant requires about 13 GB of VRAM at 8 k context.

Does this model support vision?

No, the model does not have vision capabilities.

What license does it use?

The model is released under the Apache‑2.0 license.

Other GGUFs of gpt-oss-20b-Derestricted

Similar models

Browse all →

More from Felladrin

All by Felladrin →

More gpt-oss models

All gpt-oss →

Other medium models

All medium models →