ComfyUI Node: 🦙 Ollama Image Captioner 🦙
Category
Ollama
Inputs
model
- llava:7b-v1.6-vicuna-q2_K (Q2_K, 3.2GB)
- llava:7b-v1.6-mistral-q2_K (Q2_K, 3.3GB)
- llava:7b-v1.6 (Q4_0, 4.7GB)
- llava:13b-v1.6 (Q4_0, 8.0GB)
- llava:34b-v1.6 (Q4_0, 20.0GB)
- llava-llama3:8b (Q4_K_M, 5.5GB)
- llava-phi3:3.8b (Q4_K_M, 2.9GB)
- llama3.2-vision:11b (Q4_K_M, 7.9GB)
- minicpm-v:8b (Q4_0, 5.5GB)
- moondream:1.8b (Q4, 1.7GB)
- moondream:1.8b-v2-q6_K (Q6, 2.1GB)
- moondream:1.8b-v2-fp16 (F16, 3.7GB)
custom_model STRING
api_host STRING
timeout INT
input_dir STRING
output_dir STRING
max_images INT
low_vram BOOLEAN
keep_model_alive INT
top_p FLOAT
temperature FLOAT
caption_type
- Descriptive
- Descriptive (Informal)
- FLUX/SD3+
- MidJourney
- Booru-like tag list
caption_length
- any
- very short
- short
- medium-length
- long
- very long
- 20
- 30
- 40
- 50
- 60
- 70
- 80
- 90
- 100
- 110
- 120
- 130
- 140
- 150
- 160
- 170
- 180
- 190
- 200
- 210
- 220
- 230
- 240
- 250
- 260
name STRING
custom_prompt STRING
prefix_caption STRING
suffix_caption STRING
extra_options Extra_Options
structured_output_format STRING
Outputs
STRING
Extension: ComfyUI-Ollama-Describer
A ComfyUI extension that allows you to use some LLM templates provided by Ollama, such as Gemma, Llava (multimodal), Llama2, Llama3 or Mistral
Authored by alisson-anjos
Run ComfyUI workflows in the Cloud!
No downloads or installs are required. Pay only for active GPU usage, not idle time. No complex setups and dependency issues
Learn more