ComfyUI Node
Llama.cpp Gemma 4 Runtime Preset
Outputs one typed multimodal batch configuration plus separate Gemma 4-tuned context and generation-length integers.
Llama.cpp Gemma 4 Runtime Preset
- runtime
- n_ctx
- max_tokens
◄presetVision Standard►
CategoryOllama/llama_cpp/legacy
Inputs (1)
| Name | Type | Default | Description |
|---|---|---|---|
| preset | COMBO | Vision Standard | Vision Standard is the safe starting point. Long / Thinking reserves more output context without enabling thinking. High Detail / OCR uses 1120 image tokens and may require more VRAM or backend-specific tuning. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| runtime | OLLAMA_IMAGE_LIST_LLAMA_CPP_GEMMA4_RUNTIME | — |
| n_ctx | INT | — |
| max_tokens | INT | — |