diffusers_name = diffusers_name.replace("emb.layers","time_emb_proj")# 针对 SDXL 特定的转换if"emb"indiffusers_nameand"time.emb.proj"notindiffusers_name:# 使用正则表达式移除最后一个数字pattern =r"\.\d+(?=\D*$)"diffusers_name = re.sub(pattern,"", diffusers_name, count=1)if".in."...
in_channels=self.out_channels,# 输入通道n_heads=self.num_attention_heads,# 注意力头数d_head=self.out_channels // self.num_attention_heads,# 每个头的维度depth=self.transformer_layers_per_block,# 每个块的层数use_linear_projection=self.use_linear_projection,# 是否使用线性投影only_cross_attention...
diffusers_xl_canny_full + diffusers_xl_canny_mid + diffusers_xl_canny_small + diffusers_xl_depth_full + diffusers_xl_depth_mid + diffusers_xl_depth_small 旭_1994 6枚 GPL 2 生成 0 29 2024-07-27 详情 相关项目 评论(0) 创建项目 数据集介绍 原始链接:https://hf-mirror.com/lllyasviel/...
ControlNet with Stable Diffusion XL 目前,SDXL 还没有太多配套的 ControlNet,但是 diffusers 官方已经训了两个适配 SDXL 的 ControlNet 模型,分别是以 Canny 边缘为条件的和以深度图为条件的。 这里我们以 Canny 边缘为示例。首先加载一张图片并生成其 canny 边缘图: fromdiffusersimportStableDiffusionXLControlNet...
t2i-adapter_diffusers_xl_depth_zoe.safetensors t2i-adapter_diffusers_xl_sketch.safetensors t2i-adapter_diffusers_xl_canny.safetensors t2i-adapter_diffusers_xl_lineart.safetensors t2i-adapter_diffusers_xl_openpose.safetensors t2i-adapter_diffusers_xl_depth_zoe.safetensors (150.74M) 下载关于...
load_lora_weights("CiroN2022/toy-face", weight_name="toy_face_sdxl.safetensors", adapter_name="toy") pipe.to("cuda") pipe.enable_xformers_memory_efficient_attention() res = pipe(prompt="1girl", num_inference_steps=20) Member Author sayakpaul commented Oct 24, 2023 Could you ...
SDXL Turbo is an adversarial time-distilledStable Diffusion XL(SDXL) model capable of running inference in as little as 1 step. Also, it does not use classifier-free guidance, further increasing its speed. On a good consumer GPU, you can now generate an image in just 100ms. ...
zoedepth_image_out.save('images/zoedepth_out.png') All commits 📝 Fix broken link to models documentation by @kadirnar in #4026 move to 0.19.0dev by @patrickvonplaten in #4048 [SDXL] Partial diffusion support for Text2Img and Img2Img Pipelines by @bghira in #4015 ...
SDXL-Turbo LoRA models more scheduler implementations PNDM DDIM EulerDiscrete EulerAncestralDiscrete LCM DDPM KDPM2Discrete KDPM2AncestralDiscrete more pipeline/model types txt2img img2img, inpaint, upscale, depth2img marigold async pipeline methods (sliced inference) support multiple images per prompt ...
Depth Control Canny Control Flux Redux Flux Fill Inpainting / Outpainting Flux RF-Inversion SD3.5 ControlNet ControlNet Union XL SD3.5 IP Adapter Flux IP adapter Important Note about the new Flux Models We can combine the regular Flux.1 Dev LoRAs with Flux Control LoRAs, Flux Control, and Flu...