GP

GPT OSS 26B Abliterated Preview

gpt-ossGGUF by mradermacherUpdated 2025-08-09
Download on Hugging Face
Parameters25.9B
Context128K
Layers30
Architecturegpt-oss
Smallest quant14.8 GB
Quick answer

GPT OSS 26B Abliterated Preview needs about 20.4 GB of memory at Q4_K_M (a 19.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 14.8 GB (Q3_K_S) to 27.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.0 GB.

GPT OSS 26B Abliterated Preview – a 26B parameter LLM with tools support

GPT OSS 26B Abliterated Preview is a 25.9 billion‑parameter model built on the gpt‑oss architecture. It supports tool usage and is released under the Apache‑2.0 license. The model is designed for research and controlled environments due to reduced safety filtering.

Good at

  • Tool‑enabled reasoning
  • Large context handling
  • Research‑grade language modeling

Limits

  • Reduced safety filtering – may produce sensitive content
  • No vision or audio capabilities
  • Requires careful monitoring in production

Best for

  • Controlled research experiments
  • Tool‑integrated applications
  • High‑capacity language tasks

Who it’s for. Developers and researchers who need a large, tool‑capable LLM for experimentation. Use in environments where outputs can be monitored and reviewed.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S14.8 GB 15.8 GB
Too big Get file →
Q2_K14.8 GB 15.8 GB
Too big Get file →
IQ4_XS15.0 GB 16.0 GB
Too big Get file →
Q3_K_M15.9 GB 16.9 GB
Too big Get file →
Q3_K_L16.4 GB 17.4 GB
Too big Get file →
Q4_K_S18.0 GB 19.1 GB
Too big Get file →
Q4_K_M19.4 GB 20.4 GB
Too big Get file →
Q5_K_S19.6 GB 20.6 GB
Too big Get file →
Q5_K_M20.8 GB 21.8 GB
Too big Get file →
Q6_K27.4 GB 28.4 GB
Too big Get file →
Q8_027.5 GB 28.5 GB
Too big Get file →

For 8 GB VRAM, use Q3_K_S (14.84 GB) or Q2_K (14.84 GB). For 12 GB VRAM, Q3_K_M (15.9 GB) or Q3_K_L (16.43 GB) are suitable. For 16 GB VRAM, Q4_K_S (18.05 GB) or Q4_K_M (19.4 GB) provide a balance of size and performance. Memory estimates are approximate.

What GPU do you need for GPT OSS 26B Abliterated Preview?

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 IQ4_XS (15.0 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (20.8 GB file, about 21.8 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (27.5 GB file, about 28.5 GB in use).

Which GPT OSS 26B Abliterated Preview quant should you choose?

Q4_K_M is the usual starting point for GPT OSS 26B Abliterated Preview: it balances file size and quality well. Here it is a 19.4 GB file.

The Q8_0 file is 27.5 GB, about 1.4 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 Q3_K_S at 14.8 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturegpt-oss
Parameters25.9B (approximate, mixture of experts)
Layers30
Trained context131,072 tokens
KV cache (fp16)60 KB per token

Source

Quantized bymradermacher
Familygpt-oss
Licenseapache-2.0
Downloads (HF)1,076
Last updated2025-08-09

Frequently asked questions about GPT OSS 26B Abliterated Preview

How much VRAM does GPT OSS 26B Abliterated Preview need?

About 20.4 GB at Q4_K_M with an 8K context window. That is the 19.4 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 26B Abliterated Preview on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.8 GB, so you would need a larger GPU or CPU offloading.

What is the file size of GPT OSS 26B Abliterated Preview GGUF?

It depends on the quant: Q3_K_S is 14.8 GB, Q2_K is 14.8 GB, IQ4_XS is 15.0 GB, Q3_K_M is 15.9 GB, Q3_K_L is 16.4 GB, Q4_K_S is 18.0 GB and 5 more.

What context length does GPT OSS 26B Abliterated Preview support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.5 GB at 8K and 2.0 GB at 32K.

Where can I download GPT OSS 26B Abliterated Preview GGUF?

The files are on Hugging Face in the mradermacher/GPT-OSS-26B-abliterated-Preview-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 26B Abliterated Preview 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?

Yes, the Q3_K_S and Q2_K quantizations are 14.84 GB, which fit within 8 GB VRAM when using efficient inference libraries.

Does this model support vision or audio?

No, it only handles text and tool usage.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of GPT-OSS-26B-abliterated-Preview

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More gpt-oss models

All gpt-oss →

Other medium models

All medium models →