GP

Gpt Oss 20b

gpt-ossGGUF by leeminwaanUpdated 2025-09-01
Download on Hugging Face
Parameters10.2B
Context128K
Layers24
Architecturegpt-oss
Smallest quant6.0 GB
Quick answer

Gpt Oss 20b needs about 8.6 GB of memory at Q4_K_M (a 7.7 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 15 quants, from 6.0 GB (Q3_K_S) to 20.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.

Pruned GPT-OSS 20B with 14 experts per layer

Gpt Oss 20b is a pruned variant of GPT-OSS-20B, reduced to 14 experts per layer. It is a decoder‑only transformer with 10.2B parameters and a trained context of 131072 tokens. The model is released under Apache‑2.0 and supports tools.

Good at

  • Tool‑based inference
  • Low‑memory deployment
  • Expert‑pruned efficiency

Limits

  • No vision support
  • Experimental model may have reduced performance
  • Requires 8k context for memory estimates

Best for

  • Text generation on consumer GPUs
  • Benchmarking quantization formats
  • Fine‑tuning chatbots

Who it’s for. Developers and researchers who need a smaller GPT‑OSS model for local inference. The pruning reduces memory usage while retaining a large portion of the original expert set.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S6.0 GB 6.9 GB
Fits Get file →
Q2_K6.0 GB 6.9 GB
Fits Get file →
Q4_06.0 GB 6.9 GB
Fits Get file →
Q3_K_M6.4 GB 7.3 GB
Tight Get file →
Q3_K_L6.6 GB 7.5 GB
Tight Get file →
Q4_16.6 GB 7.5 GB
Tight Get file →
Q4_K_S7.2 GB 8.1 GB
Too big Get file →
Q5_07.2 GB 8.1 GB
Too big Get file →
Q4_K_M7.7 GB 8.6 GB
Too big Get file →
Q5_K_S7.8 GB 8.7 GB
Too big Get file →
Q5_17.8 GB 8.7 GB
Too big Get file →
Q5_K_M8.3 GB 9.2 GB
Too big Get file →
Q6_K10.8 GB 11.7 GB
Too big Get file →
Q8_010.8 GB 11.7 GB
Too big Get file →
—20.3 GB 21.2 GB
Too big Get file →

For 8 GB VRAM use Q3_K_S (5.99 GB, 6.9 GB memory) or Q2_K (6 GB, 6.9 GB). For 12 GB VRAM use Q4_K_M (7.72 GB, 8.6 GB). For 16 GB VRAM use Q5_K_M (8.25 GB, 9.2 GB). Estimates are approximate.

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: the best quant that fits is Q4_1 (6.6 GB file, about 7.5 GB in use).
  • 12 GB VRAM: the best quant that fits is Q8_0 (10.8 GB file, about 11.7 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (10.8 GB file, about 11.7 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (20.3 GB file, about 21.2 GB in use).

Which Gpt Oss 20b quant should you choose?

Q4_K_M is the usual starting point for Gpt Oss 20b: it balances file size and quality well. Here it is a 7.7 GB file.

The Q8_0 file is 10.8 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 6.0 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
Parameters10.2B (approximate, mixture of experts)
Layers24
Trained context131,072 tokens
KV cache (fp16)48 KB per token

Frequently asked questions about Gpt Oss 20b

How much VRAM does Gpt Oss 20b need?

About 8.6 GB at Q4_K_M with an 8K context window. That is the 7.7 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 Q4_1: the 6.6 GB file needs about 7.5 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: Q3_K_S is 6.0 GB, Q2_K is 6.0 GB, Q4_0 is 6.0 GB, Q3_K_M is 6.4 GB, Q3_K_L is 6.6 GB, Q4_1 is 6.6 GB and 9 more.

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 leeminwaan/gpt-oss-20b-pruned-10.2B-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 this model on 8 GB VRAM?

Yes, use Q3_K_S or Q2_K quantizations, each about 6 GB and requiring roughly 6.9 GB memory at 8k context.

Does this model support image inputs?

No, the model has no vision capabilities.

What license applies?

The model is licensed under Apache-2.0.

Similar models

Browse all →

More from leeminwaan

All by leeminwaan →

More gpt-oss models

All gpt-oss →

Other medium models

All medium models →