Forked from google/gemma-4-12b
Forked from google/gemma-4-12b
Model type
Language model
3 Downloads
Capabilities
Minimum system memory
Tags
Model type
Language model
3 Downloads
Capabilities
Model type
Language model
3 Downloads
Capabilities
Minimum system memory
Tags
Model type
Language model
3 Downloads
Capabilities
Minimum system memory
Tags
Last updated
Updated 15 days agobyREADME
Gemma 4 12B's "Unified" label refers to its encoder-free design. Unlike other Gemma 4 models that use separate encoders for multimodal inputs, this model skips them entirely, projecting image patches directly into the LLM's embedding space via lightweight linear layers. All modalities feed into a single decoder-only transformer, cutting multimodal latency and enabling end-to-end fine-tuning in one pass.
The model supports context length of 256K tokens with vision capabilities
Custom Fields
Special features defined by the model author
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Parameters
Custom configuration options included with this model
Sources
The underlying model files this model uses
Minimum system memory
Tags
Last updated
Updated 15 days agobyREADME
Gemma 4 12B's "Unified" label refers to its encoder-free design. Unlike other Gemma 4 models that use separate encoders for multimodal inputs, this model skips them entirely, projecting image patches directly into the LLM's embedding space via lightweight linear layers. All modalities feed into a single decoder-only transformer, cutting multimodal latency and enabling end-to-end fine-tuning in one pass.
The model supports context length of 256K tokens with vision capabilities
Custom Fields
Special features defined by the model author
Enable Thinking
: boolean
(default=true)
Controls whether the model will think before replying
Parameters
Custom configuration options included with this model
Sources
The underlying model files this model uses
Based on
GGUF
MLX
MLX
MLX
MLX
Based on
GGUF
MLX
MLX
MLX
MLX