File size: 3,034 Bytes
f638bf2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 | [
{
"name": "Text to image (SD-Turbo, 1 step)",
"file": "txt2img-sd-turbo.json",
"note": "The smallest model: draws in one step."
},
{
"name": "Text to image (Stable Diffusion 1.5, real guidance)",
"file": "txt2img-sd15.json",
"note": "20 steps with a negative prompt; downloads 4.4 GB the first time."
},
{
"name": "Image to image",
"file": "img2img.json",
"note": "Upload a picture, pick it in Load Image, then Queue."
},
{
"name": "Hires fix (draw small, upscale the latent, refine)",
"file": "hires-fix.json",
"note": "SD-Turbo at 256 px, then 512 px."
},
{
"name": "Inpainting (redraw the masked coat)",
"file": "inpaint.json",
"note": "The coat in the example picture is already masked (transparent), so Queue redraws it from the prompt. For your own: upload a picture, right-click Load Image > Open in Mask Editor, paint, Queue."
},
{
"name": "FLUX.2 Klein text to image",
"file": "flux2-klein.json",
"note": "4B model, 4 steps; downloads 4.6 GB the first time. Needs a GPU with about 4 GB free."
},
{
"name": "FLUX 3 Klein text to image",
"file": "flux3-klein.json",
"note": "5.6B model, 16 steps, wants a full scene caption; downloads 6 GB the first time."
},
{
"name": "FLUX.2 Klein picture edit",
"file": "flux2-klein-edit.json",
"note": "Edits the picture you load, following the prompt."
},
{
"name": "Text to video (LTX-Video 2B)",
"file": "ltx-video-t2v.json",
"note": "A one-second clip (25 frames at 24 fps) from a prompt, saved as an animated WEBP. The first run downloads about 8 GB of weights."
},
{
"name": "Picture to video (LTX-Video 2B)",
"file": "ltx-video-i2v.json",
"note": "The example picture as the first frame of a one-second clip; the prompt says what happens next. The first run downloads about 8 GB of weights."
},
{
"name": "LLM writes the prompt, SD-Turbo draws it",
"file": "llm-prompt.json",
"note": "A small local language model expands your idea into a prompt."
},
{
"name": "Upscale a picture with a model",
"file": "upscale.json",
"note": "ESPCN x3, a few hundred KB."
},
{
"name": "Remove the background",
"file": "remove-background.json",
"note": "RMBG 1.4: the mask, and the cut-out on green."
},
{
"name": "Depth map",
"file": "depth.json",
"note": "Depth Anything V2 small."
},
{
"name": "Pose and object detection",
"file": "detect.json",
"note": "Boxes around what is in the picture, the list of what was found, and the body skeleton."
},
{
"name": "Speak and listen (Kokoro, then Parakeet)",
"file": "speech.json",
"note": "Text to speech (Kokoro), then the speech back to text (Parakeet)."
},
{
"name": "Image to 3D model (TRELLIS.2)",
"file": "image-to-3d.json",
"note": "Upload a picture of one object, pick it in Load Image, Queue; right-click Save 3D Model > View 3D model. About 10 GB the first time."
}
]
|