FAST & FRUGAL · 薄
Gemma
4 models from Google, living in Silvergrass Fields.
| Model | Context | Input | Output | Accepts |
|---|---|---|---|---|
| Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 262K | $0.070 | $0.230 | text, images, video |
| Gemma 3 4Bgoogle/gemma-3-4b-it | 131K | $0.040 | $0.050 | text, images |
| Gemma 3 12Bgoogle/gemma-3-12b-it | 131K | $0.050 | $0.150 | text, images |
| Gemma 3 27Bgoogle/gemma-3-27b-it | 131K | $0.080 | $0.350 | text, images |
Questions and answers
Which Gemma model has the largest context window?
Gemma 4 26B A4B (Google), with 262K tokens.
Which Gemma model is cheapest per input token?
Gemma 3 4B, at $0.040 per million input tokens ($0.050 per million output tokens).
Which Gemma models accept images?
Gemma 4 26B A4B , Gemma 3 4B, Gemma 3 12B, Gemma 3 27B (4 of 4).
Which Gemma models support reasoning?
Gemma 4 26B A4B (1 of 4).
Answers are computed from the 2026-10-24 catalog snapshot.