GP

Huihui Gpt Oss 20b Mxfp4 Abliterated v2

gpt-ossGGUF by huihui-aiUpdated 2025-09-27
Download on Hugging Face
Parameters20.9B
Context128K
Layers24
Architecturegpt-oss
Smallest quant13.8 GB
Quick answer

Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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.

20‑billion‑parameter GPT‑OSS model quantized to MXFP4

Huihui Gpt Oss 20b Mxfp4 Abliterated v2 is a 20 billion‑parameter GPT‑OSS model fine‑tuned with TRL and quantized to MXFP4. It supports a 131 072‑token context window and is released under Apache‑2.0.

Good at

  • Large‑context generation
  • Fine‑tuned instruction following
  • Research‑grade performance

Limits

  • Requires 14.7 GB VRAM at 8 k context
  • No vision or multimodal support
  • Reduced safety filtering – use with caution

Best for

  • Research prototyping
  • Controlled environment testing
  • Custom instruction fine‑tuning

Who it’s for. Researchers and developers needing a large‑context GPT‑OSS model with low memory footprint. Suitable for experimentation and controlled deployments where safety filtering can be managed.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
MXFP413.8 GB 14.7 GB
Too big Get file →

The MXFP4 quantization yields a 13.79 GB file. With 8 k context it needs about 14.7 GB VRAM. For 4 k context the VRAM requirement drops roughly by half. Use MXFP4 if you have 16 GB or more VRAM; otherwise consider a larger quant or a smaller model.

What GPU do you need for Huihui Gpt Oss 20b Mxfp4 Abliterated v2?

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 Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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

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

Source

Quantized byhuihui-ai
Familygpt-oss
Licenseapache-2.0
Downloads (HF)23,055
Last updated2025-09-27

Frequently asked questions about Huihui Gpt Oss 20b Mxfp4 Abliterated v2

How much VRAM does Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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 Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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 Huihui Gpt Oss 20b Mxfp4 Abliterated v2 GGUF?

It depends on the quant: MXFP4 is 13.8 GB.

What context length does Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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 Huihui Gpt Oss 20b Mxfp4 Abliterated v2 GGUF?

The files are on Hugging Face in the huihui-ai/Huihui-gpt-oss-20b-mxfp4-abliterated-v2 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Huihui Gpt Oss 20b Mxfp4 Abliterated v2 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 variant requires approximately 14.7 GB VRAM at 8 k context, exceeding an 8 GB limit.

Does this model support image input?

No. It is a text‑only GPT‑OSS model with no vision capabilities.

What license does it use?

Apache‑2.0, allowing free use and modification under its terms.

Other GGUFs of Huihui-gpt-oss-20b-BF16-abliterated-v2

Similar models

Browse all →

More from huihui-ai

All by huihui-ai →

More gpt-oss models

All gpt-oss →

Other medium models

All medium models →