All Models

Gemma 4 12B Instruct

gemmaTool CallingAttachmentsOpen WeightsStructured Output

Google's Gemma 4 12B Instruct is an open-weight multimodal model for text, image, audio, and video understanding, with tool calling and structured output support.

Providers1
ReleasedAug 1, 2026
Input Modalitiestext, image, video, audio
Output Modalitiestext
Tarsk Usecoding

Available Providers (1)

ProviderModel IDInput CostOutput CostContextMax OutputDocs
NanoGPTgemma-4-12b-it$0.05/MTok$0.25/MTok131.1K32.8K

Capabilities

Reasoning
Tool Calling
Attachments
Open Weights
Structured Output