Gemma 3 4B vs Gemma 3 12B
Gemma 3 4B is the cheaper option on input; Gemma 3 4B is cheaper on output by 33%. Figures below are per million tokens, from vendor pricing cited in our reporting.
| Gemma 3 4B | Gemma 3 12B | |
|---|---|---|
| Input per 1M tokens | $0.05 | $0.05 |
| Output per 1M tokens | $0.10 | $0.15 |
| Cached input per 1M | — | — |
| Context window | 131K tokens | 131K tokens |
| Parameters | — | — |
| Licence | open weights | open weights |
| Benchmark | — | — |
| Vendor | ||
| Released | — | — |
What the numbers say
- Input price gap: 0% ($0.05 vs $0.05).
- Blended three-to-one input/output rate: $0.25 for Gemma 3 4B, $0.30 for Gemma 3 12B.
- Context: 131K tokens against 131K tokens.
- Licence: Gemma 3 4B is open weights; Gemma 3 12B is open weights.