Gemma 3 12B
Google · Multimodal · Released Mar 2025
Gemma 3 12B is Google's 12-billion-parameter open-weight text model with a 131K token context window.
- Strengths
- Offers a middle ground between the smaller Gemma 3 4B and larger Gemma 3 27B, balancing capability with inference efficiency.
- Best for
- Local deployment or serverless inference when you need more reasoning capacity than a 4B model but want faster latency than a 27B variant.
- Limitations
- Smaller than Gemma 3 27B, so less suitable for highly complex reasoning or knowledge-intensive tasks that benefit from additional model capacity.
Input / 1M
$0.05
Output / 1M
$0.15
Cached input / 1M
—
Context window
131K
Price history
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 11 Jun 2026 | $0.05 | $0.15 | — | Imported from OpenRouter | openrouter.ai |