Gemma 3 12B Instruct is the 12-billion-parameter instruction-tuned model in Google DeepMind's Gemma 3 family of open-weight multimodal models. It takes text and image input and produces text output, with a 128K-token context window and support for over 140 languages, for tasks such as chat, summarization, question answering and image understanding.
Modalities
In / out price
$0.058 / $0.173per 1M
Context
131K
Max output
16K
Updated
Nov 5, 2026
Providers
4 routesCompare providers by price, privacy, and speed.
| Private | $0.058 | $0.173 | — | 131K | 16K | — | — | |
Amazon Bedrock | Private | $0.104 | $0.34 | — | 128K | 8K | — | — |
Cheaper Inference | Anonymous | $0.085 | $0.26 | $0.085 | 128K | 8K | — | — |
Surplus Intelligence | Anonymous | $0.07 | $0.21 | $0.0105 | 128K | 8K | — | — |
Privacy
2 / 4Private routing
Private tier
- Anonymous
- Private
- TEE
- E2EE
Prompt and response content is not retained after the request. Request content is visible to the inference runtime while it is being processed. Learn more
Features
3Streaming
Tool calling
Vision
Uptime
FAQ
Common questions about Gemma 3 12B Instruct.