Rent a GPU

gpt-oss 120B hardware requirements

116.8B parameters (5.1B active, mixture of experts). Needs about 71.7 GB at Q4_K_M with 8K context.

Parameters117B5.1B active (MoE)
Q4_K_M · 8K71.7 GBthe usual download
FP16 · 8K234.5 GBfull precision
Runs well on4/51GPUs and Macs we track

Memory needed at each quantization

Weights, KV cache and runtime overhead, at 8K and 32K tokens of context.

QuantQuality8K context32K context
FP16Full quality234.5 GB235.4 GB
Q8_0Near-lossless125 GB125.9 GB
Q6_KExcellent96.7 GB97.6 GB
Q5_K_MVery good84 GB84.9 GB
Q4_K_MGood, most popular71.7 GB72.6 GB
Q3_K_MNoticeably degraded58 GB58.9 GB
Q2_KHeavily degraded44.7 GB45.6 GB

Run it with Ollama

Downloads Q4_K_M by default.

ollama run gpt-oss:120b

Model card

openai/gpt-oss-120b on Hugging Face: licence, prompt format and GGUF builds.

Run it in the cloud

Run gpt-oss 120B on a rented GPU today.

It needs about 71.7 GB at Q4_K_M: a rented A100 / H100 80 GB runs it for about $1.11 to $3.99 an hour.

RunPodRent a GPU on RunPodVast.aiRent a GPU on Vast.ai

Every GPU and Mac checked

Sorted by memory. Bars show how much of each card gpt-oss 120B fills at the best quant that fits.

4 Runs well

NameMemory usedBest quantSpeed

3 Fits with heavy quantization

NameMemory usedBest quantSpeed

44 Too big

NameMemory usedBest quantSpeed

Other gpt-oss models