The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Refactor.
comfyanonymousfcf513e0b6b5
diff --git a/comfy/sd.py b/comfy/sd.pyindex d60b908b8..174ed35e5 100644--- a/comfy/sd.py+++ b/comfy/sd.py@@ -515,6 +515,8 @@ class VAE:def decode_tiled_(self, samples, tile_x=64, tile_y=64, overlap = 16):steps = samples.shape[0] * utils.get_tiled_scale_steps(samples.shape[3], samples.shape[2], tile_x, tile_y, overlap)+ steps += samples.shape[0] * utils.get_tiled_scale_steps(samples.shape[3], samples.shape[2], tile_x // 2, tile_y * 2, overlap)+ steps += samples.shape[0] * utils.get_tiled_scale_steps(samples.shape[3], samples.shape[2], tile_x * 2, tile_y // 2, overlap)pbar = utils.ProgressBar(steps)decode_fn = lambda a: (self.first_stage_model.decode(1. / self.scale_factor * a.to(self.device)) + 1.0)@@ -566,7 +568,9 @@ class VAE:self.first_stage_model = self.first_stage_model.to(self.device)pixel_samples = pixel_samples.movedim(-1,1).to(self.device)- steps = utils.get_tiled_scale_steps(pixel_samples.shape[3], pixel_samples.shape[2], tile_x, tile_y, overlap)+ steps = pixel_samples.shape[0] * utils.get_tiled_scale_steps(pixel_samples.shape[3], pixel_samples.shape[2], tile_x, tile_y, overlap)+ steps += pixel_samples.shape[0] * utils.get_tiled_scale_steps(pixel_samples.shape[3], pixel_samples.shape[2], tile_x // 2, tile_y * 2, overlap)+ steps += pixel_samples.shape[0] * utils.get_tiled_scale_steps(pixel_samples.shape[3], pixel_samples.shape[2], tile_x * 2, tile_y // 2, overlap)pbar = utils.ProgressBar(steps)samples = utils.tiled_scale(pixel_samples, lambda a: self.first_stage_model.encode(2. * a - 1.).sample() * self.scale_factor, tile_x, tile_y, overlap, upscale_amount = (1/8), out_channels=4, pbar=pbar)diff --git a/comfy/utils.py b/comfy/utils.pyindex 5c7143fd9..09e05d4ed 100644--- a/comfy/utils.py+++ b/comfy/utils.py@@ -1,4 +1,5 @@import torch+import mathdef load_torch_file(ckpt, safe_load=False):if ckpt.lower().endswith(".safetensors"):@@ -63,10 +64,7 @@ def common_upscale(samples, width, height, upscale_method, crop):return torch.nn.functional.interpolate(s, size=(height, width), mode=upscale_method)def get_tiled_scale_steps(width, height, tile_x, tile_y, overlap):- it_1 = -(height // -(tile_y * 2 - overlap)) * -(width // -(tile_x // 2 - overlap))- it_2 = -(height // -(tile_y // 2 - overlap)) * -(width // -(tile_x * 2 - overlap))- it_3 = -(height // -(tile_y - overlap)) * -(width // -(tile_x - overlap))- return it_1 + it_2 + it_3+ return math.ceil((height / (tile_y - overlap))) * math.ceil((width / (tile_x - overlap)))@torch.inference_mode()def tiled_scale(samples, function, tile_x=64, tile_y=64, overlap = 8, upscale_amount = 4, out_channels = 3, pbar = None):diff --git a/comfy_extras/nodes_upscale_model.py b/comfy_extras/nodes_upscale_model.pyindex f774b4b77..ab5b0ccfc 100644--- a/comfy_extras/nodes_upscale_model.py+++ b/comfy_extras/nodes_upscale_model.py@@ -40,7 +40,7 @@ class ImageUpscaleWithModel:tile = 128 + 64overlap = 8- steps = -(in_img.shape[2] // -(tile - overlap)) * -(in_img.shape[3] // -(tile - overlap))+ steps = in_img.shape[0] * comfy.utils.get_tiled_scale_steps(in_img.shape[3], in_img.shape[2], tile_x=tile, tile_y=tile, overlap=overlap)pbar = comfy.utils.ProgressBar(steps)s = comfy.utils.tiled_scale(in_img, lambda a: upscale_model(a), tile_x=tile, tile_y=tile, overlap=overlap, upscale_amount=upscale_model.scale, pbar=pbar)upscale_model.cpu()