ComfyUI Node
SDNQ Sampler V2
Generate images using SDNQ models. Connect SDNQ Model Loader to provide the model.
SDNQ Sampler V2
- model
- latent_image
- image
◄prompt►
◄negative_promptblurry, low quality, distorted, deformed, ugly, bad anatomy, bad hands, text, watermark, signature►
◄steps25►
◄cfg7.0►
◄seed0►
◄schedulerDPMSolverMultistepScheduler►
◄denoise1.00►
Categorysampling/SDNQ
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | SDNQ model pipeline from SDNQ Model Loader node | |
| prompt | STRING | Text description of the image to generate. Be descriptive for best results. | |
| negative_prompt | STRING | blurry, low quality, distorted, deformed, ugly, bad anatomy, bad hands, text, watermark, signature | What to avoid in the image. Default includes common quality issues. Clear this for no negative prompt. Note: FLUX-schnell (cfg=0) ignores negative prompts. |
| steps | INT | 251–150 | Number of denoising steps. More steps = better quality but slower. 20-30 is typical for most models. |
| cfg | FLOAT | 7.00–30 | Guidance scale - how closely to follow the prompt. Higher = more literal. FLUX-schnell uses 0.0, FLUX-dev uses 3.5-7.0, SDXL uses 7.0-9.0. |
| latent_image | LATENT | Latent image input. Width and height are extracted from this latent tensor. | |
| seed | INT | 00–18446744073709550000 | The random seed used for creating the noise. Seed=0 will randomize each run and update this field after generation. |
| scheduler | COMBO | DPMSolverMultistepScheduler | ⚠️ IMPORTANT: Use FlowMatchEulerDiscreteScheduler for FLUX/SD3/Qwen/Z-Image. Use DPMSolver/Euler/UniPC for SDXL/SD1.5. Wrong scheduler = broken images! |
| denoise | FLOAT | 1.000–1 | Denoising strength. 1.0 = full denoising, lower values = less denoising. ComfyUI standard parameter matching KSampler. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |