CanItRun Logocanitrun.

GPT-OSS 20B

GPT-OSS 20B needs roughly 14.8 GB VRAM at Q4_K_M quantization (47.5 GB at FP16). 84 GPUs we track can run it fully in VRAM at 8k context.

84 GPUs run this natively · 11 with CPU offload

OpenAI21B params4B active (MoE)128k contextApache 2.0Commercial use ok

GPT-OSS 20B is a Mixture of Experts (MoE) model with 21B total parameters but only 4B active per token developed by OpenAI. August 2025 21B MoE with 4B active — matches o3-mini on key benchmarks.

To run GPT-OSS 20B locally: Q5_K_M ~14-16GB — fits on 16GB GPUs. Best reasoning model for 16GB hardware. As a MoE model, inference speed depends on active parameters (4B) rather than total size.

GPQA 71.5% at 21B scale — exceptional reasoning efficiency.

VRAM at each quantization

Figures below assume 8k context; KV cache grows linearly as context length increases.

QuantWeightsKV cacheTotal
FP3284.0 GB0.40 GB94.5 GB
BF1642.0 GB0.40 GB47.5 GB
FP1642.0 GB0.40 GB47.5 GB
Q8_022.3 GB0.40 GB25.4 GB
Q6_K17.2 GB0.40 GB19.8 GB
Q5_K_Mrec14.9 GB0.40 GB17.2 GB
Q4_K_M12.8 GB0.40 GB14.8 GB
Q3_K_M10.1 GB0.40 GB11.8 GB
Q2_K8.0 GB0.40 GB9.4 GB
NVFP4cuda10.5 GB0.40 GB12.2 GB

Shown at 8k context with FP16 KV cache. NVFP4 needs a CUDA GPU to run. Toggle TurboQuant in the calculator to view compressed KV cache numbers.

Benchmarks

MMLU-Pro
67.86
GPQA Diamond
71.5

GPUs that run GPT-OSS 20B natively (84)

Plus 11 GPUs that run it with CPU offload (slower)

Notes

Smaller sibling of GPT-OSS 120B. Matches o3-mini on key benchmarks; runs on 16 GB of VRAM.

Hugging Face ↗Released 2025-08-05

Explore

Browse all guides →Browse all models →Browse all GPUs →

Frequently asked questions

What are the VRAM requirements for GPT-OSS 20B?
GPT-OSS 20B requires approximately 14.8 GB of VRAM at Q4_K_M quantization, 25.5 GB at Q8, and 47.5 GB at FP16. These numbers assume 8k context window; VRAM scales linearly with context length due to the KV cache.
How many parameters does GPT-OSS 20B have?
GPT-OSS 20B has 21 billion total parameters, but only 4 billion are active per token thanks to its Mixture of Experts (MoE) architecture. This makes inference significantly faster than the total parameter count suggests.
How capable is GPT-OSS 20B?
With an MMLU-Pro score of 67.86, GPT-OSS 20B delivers solid general-purpose performance suitable for most everyday tasks and professional use.
Can GPT-OSS 20B run on a 16 GB GPU?
Yes. GPT-OSS 20B needs 14.8 GB at Q4_K_M, which fits in a 16 GB GPU like the RTX 4080 or RTX 4070 Ti Super.
What is the smallest quantization for GPT-OSS 20B that fits in 24 GB of VRAM?
At NVFP4, GPT-OSS 20B needs 12.2 GB — the highest-quality quantization that fits in 24 GB of VRAM.
What GPU do I need to run GPT-OSS 20B locally?
A 16 GB GPU is enough. At Q4_K_M, GPT-OSS 20B needs 14.8 GB VRAM. Good options: RTX 4080 (16 GB), RTX 4070 Ti Super (16 GB).