Rent a GPU

Gemma 3 1B hardware requirements

1B parameters. Needs about 1.3 GB at Q4_K_M with 8K context.

Parameters1BDense
Q4_K_M · 8K1.3 GBthe usual download
FP16 · 8K2.6 GBfull precision
Runs well on51/51GPUs 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 quality2.6 GB2.7 GB
Q8_0Near-lossless1.7 GB1.8 GB
Q6_KExcellent1.5 GB1.6 GB
Q5_K_MVery good1.4 GB1.5 GB
Q4_K_MGood, most popular1.3 GB1.4 GB
Q3_K_MNoticeably degraded1.1 GB1.2 GB
Q2_KHeavily degraded1 GB1.1 GB

Run it with Ollama

Downloads Q4_K_M by default.

ollama run gemma3:1b

Model card

google/gemma-3-1b-it on Hugging Face: licence, prompt format and GGUF builds.

Run it in the cloud

Run Gemma 3 1B on a rented GPU today.

It needs about 1.3 GB at Q4_K_M: a rented RTX 3090 / RTX 4090 runs it for about $0.18 to $0.89 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 Gemma 3 1B fills at the best quant that fits.

51 Runs well

NameMemory usedBest quantSpeed

Other Gemma models