Gemma 3 models
13 Gemma 3 models on Hugging Face, from 268M to 27.4B parameters. At the precision each one is published in, the smallest needs about 0.6 GB of VRAM (gemma-3-270m, cheapest live fit: RTX 5060 Ti) and the largest about 61.3 GB (gemma-3-27b-it, cheapest live fit: A100). The cheapest way to run gemma-3-270m is $0.110/hr.
Gemma 3 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| gemma-3-270m | 268M | BF16 | 0.6 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-270m-it | 268M | BF16 | 0.6 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-270m-it | 268M | BF16 | 0.6 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-1b-it | 1000M | BF16 | 2.2 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-1b-pt | 1000M | BF16 | 2.2 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-1b-it | 1000M | BF16 | 2.2 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-4B-T1-it | 3.9B | BF16 | 8.7 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-4b-it | 4.3B | BF16 | 9.6 GB | RTX 5060 Ti | $0.110/hr |
| gemma-3-12b-it-heretic | 12.2B | BF16 | 27.2 GB | RTX 4080 Super | $0.338/hr |
| gemma-3-12b-it-heretic-v2 | 12.2B | BF16 | 27.2 GB | RTX 4080 Super | $0.338/hr |
| gemma-3-12b-it | 12.2B | BF16 | 27.2 GB | RTX 4080 Super | $0.338/hr |
| gemma-3-27b-it | 27.4B | BF16 | 61.3 GB | A100 | $1.21/hr |
| gemma-3-27b-it-FP8-dynamic | 27.4B | F8_E4M3 | 30.7 GB | RTX 4080 Super | $0.338/hr |
VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; see the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
