| Spec | Gemma 4 31B | Gemma 4 12B |
|---|---|---|
| Lowest input | $0.12Doubleword | $0.30Sail Research |
| Lowest output | $0.35Doubleword | $2.00Sail Research |
| Lowest cached input | $0.12Doubleword | $0.15Sail Research |
| Context | 262K | 16.4K |
| Max output | 131K | 16.4K |
| Providers | 4 | 1 |
| Released | April 2, 2026 | May 23, 2026 |
| Weights | Open | Open |
| Lab | ||
| Model ID | google/gemma-4-31b-it | google/gemma-4-12b-it |
USD per 1M tokens, as of November 5, 2026.
| Provider | Gemma 4 31B | Gemma 4 12B |
|---|---|---|
| Sail Research | $0.40 / $0.60 | $0.30 / $2.00 |
| DeepInfra | $0.20 / $0.40 | – |
| Doubleword | $0.12 / $0.35 | – |
| Novita AI | $0.14 / $0.40 | – |
Input / output, USD per 1M tokens.
| Model | 1,000 chat replies2K in, 500 out each | 1,000 agent tasks50K in, 5K out each | Monthly batch job500M in, 50M out |
|---|---|---|---|
| Gemma 4 31BDoubleword | $0.42 | $7.75 | $77.50 |
| Gemma 4 12BSail Research | $1.60 | $25.00 | $250.00 |
On each model's cheapest provider by input plus output list price.
Input: Gemma 4 31B at $0.12 per million input tokens vs $0.30 (Gemma 4 12B), 2.5x lower. Output: Gemma 4 31B at $0.35 per million output tokens vs $2.00 (Gemma 4 12B), 5.7x lower. Lowest listed prices as of November 5, 2026.
Gemma 4 31B, with 262K tokens vs 16.4K.
1 provider serves both: Sail Research. Gemma 4 31B has 4 providers in total and Gemma 4 12B has 1.
Gemma 4 31B from $0.12 per million cached input tokens (Doubleword). Gemma 4 12B from $0.15 (Sail Research).
Yes. Both use the same OpenAI-compatible endpoint, https://api.routerplus.com/v1, and key. Change the model field from "google/gemma-4-31b-it" to "google/gemma-4-12b-it".
