ComfyUI Node
Flamin Galah Image Describer
Flamin Galah Image Describer – local Ollama only. Vision model describes the image; prompt model writes I2V (H3 image-to-video fields) or I2I (plain paragraph).
Flamin Galah Image Describer
- image
- h3_prompt
◄output_modeImage to Video►
◄extra_description►
◄ollama_urlhttp://localhost:11434►
◄vision_model(Ollama not running – start it and refresh)►
◄prompt_model(Ollama not running – start it and refresh)►
◄temperature0.20►
◄lora_tag►
Categoryprompt/Flamin Galah
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | — | |
| output_mode | COMBO | Image to Video | 2 options: Image to Video, Image to Image |
| extra_description | STRING | — | |
| ollama_urlopt | STRING | http://localhost:11434 | — |
| vision_modelopt | COMBO | (Ollama not running – start it and refresh) | 11 options: (Ollama not running – start it and refresh), llava, llava:13b, qwen2.5-vl, llama3.2-vision, llama3.2, +5 |
| prompt_modelopt | COMBO | (Ollama not running – start it and refresh) | 11 options: (Ollama not running – start it and refresh), llava, llava:13b, qwen2.5-vl, llama3.2-vision, llama3.2, +5 |
| temperatureopt | FLOAT | 0.200–1.5 | — |
| lora_tagopt | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| h3_prompt | STRING | — |