ComfyUI Node
Context-Anchored Tile Upscale (VL)
A ComfyUI node in image/upscaling with 19 inputs and 1 output.
Context-Anchored Tile Upscale (VL)
- image
- model
- clip
- vae
- upscale_model
- negative
- IMAGE
◄seed0►
◄sampler_namedpmpp_2m►
◄schedulersgm_uniform►
◄steps20►
◄cfg3.5►
◄denoise0.50►
◄upscale_by2.00►
◄max_tile_width1536►
◄max_tile_height2048►
◄context_anchor32►
◄context_overlap32►
◄anchor_sourcesource image►
◄vlm_methodvision tokens and captions►
Categoryimage/upscaling
Inputs (19)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | The image to upscale and then refine. | |
| model | MODEL | The diffusion model that denoises each tile. | |
| clip | CLIP | Must be a vision-language text encoder (Krea 2 family). There is no positive prompt input, since each tile is conditioned on the image itself. | |
| vae | VAE | The VAE that encodes and decodes each tile. | |
| seed | INT | 00–18446744073709550000 | Noise is drawn once for the entire image and then sliced for each tile. |
| sampler_name | COMBO | dpmpp_2m | 44 options: euler, euler_cfg_pp, euler_ancestral, euler_ancestral_cfg_pp, heun, heunpp2, +38 |
| scheduler | COMBO | sgm_uniform | 9 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +3 |
| steps | INT | 201–10000 | — |
| cfg | FLOAT | 3.50–100 | — |
| denoise | FLOAT | 0.500–1 | — |
| upscale_by | FLOAT | 2.000.01–8 | The upscale multiplier. The optional upscale_model runs first when one is connected. |
| max_tile_width | INT | 1536256–16384 | Hard cap on the width the model ever sees per sampled crop, including the context_overlap and context_anchor rings. Set to the largest width the model supports. |
| max_tile_height | INT | 2048256–16384 | Hard cap on the height the model ever sees per sampled crop, including the context_overlap and context_anchor rings. Set to the largest height the model supports. |
| context_anchor | INT | 320–512 | Pixels around each tile that are frozen and shown to the model as context, then cropped away. |
| context_overlap | INT | 320–512 | Overlapped context that is diffused from both sides and then blended. It anchors the tiles to each other, like context_anchor anchors each tile to its surroundings. |
| anchor_source | COMBO | source image | What fills context_anchor. source image keeps the result true to the input. live canvas adds more detail and drifts further from the input. |
| vlm_method | COMBO | vision tokens and captions | Whether each tile is conditioned on a caption of itself, on its slice of the entire image's vision encode, or on both. The name in parentheses is the caption preset it asks. Copy settings.toml to settings.user.toml to write your own tile prompts. |
| upscale_modelopt | UPSCALE_MODEL | Optional upscale model, run over the entire image before any tiling. | |
| negativeopt | CONDITIONING | Optional negative conditioning. Unconnected it is an empty encode of this node's CLIP. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| IMAGE | IMAGE | — |