GP

gpt-oss-20b

gpt-ossGGUF by ggml-orgUpdated 2026-10-19
Download on Hugging Face
Parameters870M
Context128K
Layers1
Architectureeagle3
Smallest quant0.9 GB
Quick answer

gpt-oss-20b needs about 2.2 GB of memory at BF16 (a 1.7 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.9 GB (Q8_0) to 12.1 GB (MXFP4). Longer context needs more: at 32K tokens the KV cache alone adds about 0.1 GB.

gpt-oss-20b: 0.87B parameter open-weight model for reasoning and agentic tasks

gpt-oss-20b is a 0.87B parameter model built on the eagle3 architecture. It supports full chain-of-thought reasoning and agentic capabilities such as function calling and code execution. The model is released under an Apache‑2.0 license.

Good at

  • Reasoning with chain‑of‑thought
  • Agentic tasks like function calling
  • Fine‑tuning for custom use cases

Limits

  • Requires 8k context for memory estimates
  • No vision support
  • Large memory for higher precision quants

Best for

  • Local inference on 16GB GPUs
  • Low‑latency applications
  • Fine‑tuning for specialized domains

Who it’s for. Developers needing an open‑weight model for reasoning or agentic tasks. Researchers and businesses can customize the model under a permissive license.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q8_00.9 GB 1.4 GB
Fits Get file →
BF161.7 GB 2.2 GB
Fits Get file →
MXFP412.1 GB 12.6 GB
Too big Get file →

For 8GB VRAM, use Q8_0 (0.92 GB file, ~1.4 GB memory at 8k context). With 16GB VRAM, MXFP4 (12.11 GB file, ~12.6 GB memory) is suitable. BF16 requires ~2.2 GB memory and 1.72 GB file size.

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: the best quant that fits is BF16 (1.7 GB file, about 2.2 GB in use).
  • 6 GB VRAM: the best quant that fits is BF16 (1.7 GB file, about 2.2 GB in use).
  • 8 GB VRAM: the best quant that fits is BF16 (1.7 GB file, about 2.2 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (1.7 GB file, about 2.2 GB in use).
  • 16 GB VRAM: the best quant that fits is MXFP4 (12.1 GB file, about 12.6 GB in use).

Which gpt-oss-20b quant should you choose?

BF16 (1.7 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.

If memory is tight, the smallest file is Q8_0 at 0.9 GB, with the quality trade-offs that come with fewer bits. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureeagle3
Parameters870M
Layers1
Trained context131,072 tokens
KV cache (fp16)2 KB per token

Source

Quantized byggml-org
Familygpt-oss
Licenseapache-2.0
Downloads (HF)253,688
Last updated2026-10-19

Frequently asked questions about gpt-oss-20b

How much VRAM does gpt-oss-20b need?

About 2.2 GB at BF16 with an 8K context window. That is the 1.7 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?

Yes, with BF16: the 1.7 GB file needs about 2.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of gpt-oss-20b GGUF?

It depends on the quant: Q8_0 is 0.9 GB, BF16 is 1.7 GB, MXFP4 is 12.1 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 0.0 GB at 8K and 0.1 GB at 32K.

Where can I download gpt-oss-20b GGUF?

The files are on Hugging Face in the ggml-org/gpt-oss-20b-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 on 8 GB VRAM?

Yes, the Q8_0 quant uses 0.92 GB file size and estimates 1.4 GB memory at 8k context.

Is the model fine‑tunable?

Yes, it supports full fine‑tuning for custom use cases.

Other GGUFs of gpt-oss-20b

Similar models

Browse all →

More from ggml-org

All by ggml-org →

More gpt-oss models

All gpt-oss →

Other tiny models

All tiny models →