ComfyUI Node
Image Upscale Sampler (CRT)
A ComfyUI node in CRT/Sampling with 35 inputs and 2 outputs.
Image Upscale Sampler (CRT)
- model
- positive
- negative
- image
- latent
- vae
- image
- latent
◄sampler_nameeuler►
◄schedulersimple►
◄cfg1.0►
◄steps4►
◄denoise0.33►
◄seed0►
◄seed_shift0►
◄edit_model_flux2kleinfalse►
◄enable_upscale_modelfalse►
◄upscale_model_name►
◄tile_size_megapixels1.0►
◄tile_grid4x4►
◄tile_padding5.0►
◄mask_blur10.0►
◄stage1_sigma_factor1.005►
◄stage2_sigma_factor0.995►
◄stage1_sigma_start0.300►
◄stage1_sigma_end0.500►
◄stage2_sigma_start0.000►
◄stage2_sigma_end0.700►
◄details_amount_stage10.50►
◄details_amount_stage20.15►
◄enable_noise_injectiondisable►
◄injection_point0.75►
◄injection_seed_offset1►
◄injection_strength0.25►
◄normalize_injected_noiseenable►
◄vae_decode_tiledtrue►
◄color_match_strength1.00►
CategoryCRT/Sampling
Inputs (35)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | — | |
| positive | CONDITIONING | — | |
| sampler_name | COMBO | euler | 44 options: euler, euler_cfg_pp, euler_ancestral, euler_ancestral_cfg_pp, heun, heunpp2, +38 |
| scheduler | COMBO | simple | 9 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +3 |
| cfg | FLOAT | 1.00–100 | Classifier Free Guidance scale. Higher values follow the prompt more closely. |
| steps | INT | 41–10000 | — |
| denoise | FLOAT | 0.330–1 | Amount of denoising to apply. 1.0 = full denoising (txt2img), 0.5-0.8 typical for img2img. |
| seed | INT | 00–18446744073709550000 | — |
| seed_shift | INT | 0-100000–100000 | Offset added to the main seed for variation |
| edit_model_flux2klein | BOOLEAN | false | Enable per-tile ReferenceLatent-style conditioning for edit models |
| enable_upscale_model | BOOLEAN | false | Enable AI model upscaling before sampling |
| upscale_model_name | COMBO | Select upscale model from models/upscale_models folder | |
| tile_size_megapixels | FLOAT | 1.00.1–16 | Target megapixels per tile used to derive upscale factor |
| tile_grid | COMBO | 4x4 | Grid size for tiling (e.g., 4x4 = 16 tiles) |
| tile_padding | FLOAT | 5.00–25 | Padding around each tile as percentage of tile size |
| mask_blur | FLOAT | 10.00–50 | Blur radius for tile blending as percentage of tile size |
| stage1_sigma_factor | FLOAT | 1.0050–100 | — |
| stage2_sigma_factor | FLOAT | 0.9950–100 | — |
| stage1_sigma_start | FLOAT | 0.3000–1 | — |
| stage1_sigma_end | FLOAT | 0.5000–1 | — |
| stage2_sigma_start | FLOAT | 0.0000–1 | — |
| stage2_sigma_end | FLOAT | 0.7000–1 | — |
| details_amount_stage1 | FLOAT | 0.50-5–5 | — |
| details_amount_stage2 | FLOAT | 0.15-5–5 | — |
| enable_noise_injection | COMBO | disable | Enable noise injection during sampling |
| injection_point | FLOAT | 0.750–1 | Percentage of steps after which to inject noise |
| injection_seed_offset | INT | 1-100–100 | Offset added to main seed for injection noise |
| injection_strength | FLOAT | 0.25-20–20 | Strength of injected noise |
| normalize_injected_noise | COMBO | enable | Normalize injected noise to match latent statistics |
| vae_decode_tiled | BOOLEAN | true | Use tiled VAE decode. Tile size and overlap are derived from tile size megapixels and tile padding. |
| color_match_strength | FLOAT | 1.000–2 | Strength of color matching between original and enhanced result. 0.0 = disabled, 1.0 = full matching |
| negativeopt | CONDITIONING | — | |
| imageopt | IMAGE | — | |
| latentopt | LATENT | — | |
| vaeopt | VAE | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| latent | LATENT | — |