ComfyUI Node
Prompt Generator Options
Configure LLM model, sampling parameters, and system prompt for the Prompt Generator.
Prompt Generator Options
- image2
- image3
- image4
- image5
- options
◄modelQwen3.5-9B-UD-Q4_K_XL.gguf►
◄system_prompt_modeappend►
◄system_prompt►
◄use_model_default_samplingtrue►
◄temperature0.70►
◄top_k20►
◄top_p0.90►
◄min_p0.05►
◄repeat_penalty1.00►
◄context_size4096►
◄show_everything_in_consolefalse►
CategoryPrompt Manager
Inputs (15)
| Name | Type | Default | Description |
|---|---|---|---|
| modelopt | COMBO | Qwen3.5-9B-UD-Q4_K_XL.gguf | Select model to use (local models listed first, then HuggingFace models) Download sizes: UD-Q4_K_XL ~6GB | Q8_0 ~9.5GB | UD-Q8_K_XL ~13GB |
| image2opt | IMAGE | Connect an image (required for 'Analyze Image' and 'Analyze Image with Prompt' modes) | |
| image3opt | IMAGE | Connect an image (required for 'Analyze Image' and 'Analyze Image with Prompt' modes) | |
| image4opt | IMAGE | Connect an image (required for 'Analyze Image' and 'Analyze Image with Prompt' modes) | |
| image5opt | IMAGE | Connect an image (required for 'Analyze Image' and 'Analyze Image with Prompt' modes) | |
| system_prompt_modeopt | COMBO | append | replace: the text below fully replaces the default LLM instructions append: the text below is added after the default LLM instructions |
| system_promptopt | STRING | Custom LLM Instructions (leave empty to use default) The default instructions are designed for generating detailed and imaginative prompts for text-to-image generation. | |
| use_model_default_samplingopt | BOOLEAN | true | Use the model's default sampling parameters (overrides temperature, top_p, etc) |
| temperatureopt | FLOAT | 0.700–2 | Controls randomness (0.0 = deterministic, 2.0 = very random) |
| top_kopt | INT | 200–100 | Sample from top K most likely tokens (0 = disabled) |
| top_popt | FLOAT | 0.900–1 | Nucleus sampling: consider tokens with top_p probability mass |
| min_popt | FLOAT | 0.050–1 | Minimum probability threshold relative to top token |
| repeat_penaltyopt | FLOAT | 1.001–2 | Penalty for repeating tokens (1.0 = no penalty) |
| context_sizeopt | INT | 4096512–32768 | Context size (increase for vision models or large prompts) |
| show_everything_in_consoleopt | BOOLEAN | false | Print system prompt, user prompt, thinking process, and raw model response to console |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| options | OPTIONS | — |