comfyanonymous / ComfyUI

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
https://www.comfy.org/
GNU General Public License v3.0
58.75k stars 6.23k forks source link

The prompt that runs on your own server with the websockets_api_example.py code returns an empty result #5064

Closed DFires closed 4 weeks ago

DFires commented 2 months ago

Your question

image

Using websockets_api_example.py code to run the prompt on your own server returns an empty result, and the same prompt json file can be used manually to run the web page in the same environment to generate images

this is my prompt: { "22": { "class_type": "PulidModelLoader", "inputs": { "pulid_file": "ip-adapter_pulid_sdxl_fp16.safetensors" }, "_meta": { "title": "PuLID模型加载器" } }, "44": { "class_type": "InstantIDModelLoader", "inputs": { "instantid_file": "ip-adapter.bin" }, "_meta": { "title": "InstnatID模型加载器" } }, "88": { "class_type": "UpscaleModelLoader", "inputs": { "model_name": "4x_NMKD-Siax_200k.pth" }, "_meta": { "title": "放大模型加载器" } }, "23": { "class_type": "PulidEvaClipLoader", "inputs": {}, "_meta": { "title": "PuLIDEVAClip加载器" } }, "45": { "class_type": "InstantIDFaceAnalysis", "inputs": { "provider": "CPU" }, "_meta": { "title": "InstantID面部分析" } }, "89": { "class_type": "ImageScaleBy", "inputs": { "image": [ "87", 0 ], "scale_by": 0.38, "upscale_method": "nearest-exact" }, "_meta": { "title": "图像按系数缩放" } }, "46": { "class_type": "ControlNetLoader", "inputs": { "control_net_name": "diffusion_pytorch_model.safetensors" }, "_meta": { "title": "ControlNet加载器" } }, "47": { "class_type": "KSampler (Efficient)", "inputs": { "vae_decode": "true", "latent_image": [ "90", 0 ], "seed": [ "93", 0 ], "cfg": 5, "positive": [ "43", 1 ], "steps": 25, "script": [ "75", 0 ], "scheduler": "karras", "negative": [ "43", 2 ], "denoise": 0.58, "sampler_name": "dpmpp_2m_sde", "optional_vae": [ "11", 4 ], "preview_method": "none", "model": [ "43", 0 ] }, "_meta": { "title": "K采样器(效率)" } }, "29": { "class_type": "ApplyPulid", "inputs": { "end_at": 1, "image": [ "86", 0 ], "pulid": [ "22", 0 ], "method": "fidelity", "weight": 0.3, "model": [ "10", 0 ], "eva_clip": [ "23", 0 ], "face_analysis": [ "76", 0 ], "start_at": 0 }, "_meta": { "title": "应用PuLID" } }, "90": { "class_type": "VAEEncode", "inputs": { "pixels": [ "89", 0 ], "vae": [ "11", 4 ] }, "_meta": { "title": "VAE编码" } }, "92": { "class_type": "String Literal", "inputs": { "string": "professional,4k,highly detailed,medium portrait soft light, beautiful model,vivid,photorealistic, vivid colors, ", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "93": { "class_type": "Seed Generator", "inputs": { "seed": 1047523952660796 }, "_meta": { "title": "Seed Generator" } }, "94": { "class_type": "String Literal", "inputs": { "string": "1man,indoors,business suit,necktie,,upper_body,portrait, ", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "75": { "class_type": "Noise Control Script", "inputs": { "seed": 813662192183951, "cfg_denoiser": false, "add_seed_noise": false, "weight": 0.015, "rng_source": "gpu" }, "_meta": { "title": "控制噪波" } }, "10": { "class_type": "Efficient Loader", "inputs": { "lora_name": "None", "weight_interpretation": "A1111", "batch_size": 2, "empty_latent_width": 832, "empty_latent_height": 1216, "speak_and_recognation": true, "token_normalization": "mean", "lora_stack": [ "103", 0 ], "positive": [ "99", 0 ], "lora_clip_strength": 1, "ckpt_name": "leosamsHelloworldXL_helloworldXL70.safetensors", "negative": "(worst quality, low quality, blurry, bad eye, wrong hand, bad anatomy, wrong anatomy, open mouth, deformed, distorted, disfigured, cgi, illustration, cartoon, poorly drawn, watermark),nsfw,nipples,flag,american_flag, ", "lora_model_strength": 1, "clip_skip": -1, "vae_name": "Baked VAE" }, "_meta": { "title": "效率加载器" } }, "76": { "class_type": "PulidInsightFaceLoader", "inputs": { "provider": "CPU" }, "_meta": { "title": "PuLIDInsightFace加载器" } }, "11": { "class_type": "KSampler (Efficient)", "inputs": { "vae_decode": "true", "latent_image": [ "10", 3 ], "seed": [ "93", 0 ], "cfg": 6, "positive": [ "10", 1 ], "steps": 25, "script": [ "75", 0 ], "scheduler": "karras", "negative": [ "10", 2 ], "denoise": 1, "sampler_name": "dpmpp_2m_sde", "optional_vae": [ "10", 4 ], "preview_method": "none", "model": [ "29", 0 ] }, "_meta": { "title": "K采样器(效率)" } }, "99": { "class_type": "Text Concatenate", "inputs": { "delimiter": ", ", "text_a": [ "94", 0 ], "clean_whitespace": "true", "text_c": [ "92", 0 ], "text_b": [ "100", 0 ] }, "_meta": { "title": "文本连锁" } }, "16": { "class_type": "LoadImage", "inputs": { "image": "94a506e555b24c20969b19a401825986.png", "upload": "image" }, "_meta": { "title": "加载图像" } }, "100": { "class_type": "String Literal", "inputs": { "string": "", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "103": { "class_type": "easy loraStack", "inputs": { "lora_3_name": "None", "lora_6_name": "None", "lora_7_strength": 1, "lora_9_clip_strength": 1, "lora_1_strength": 0.6, "lora_9_model_strength": 1, "lora_5_model_strength": 1, "lora_4_clip_strength": 1, "lora_2_clip_strength": 1, "lora_4_strength": 1, "mode": "simple", "num_loras": 1, "lora_6_clip_strength": 1, "lora_8_name": "None", "lora_7_name": "None", "lora_8_model_strength": 1, "lora_5_strength": 1, "lora_4_name": "None", "lora_4_model_strength": 1, "lora_1_name": "None", "lora_2_strength": 1, "lora_2_model_strength": 1, "lora_3_clip_strength": 1, "lora_8_clip_strength": 1, "lora_10_model_strength": 1, "lora_7_model_strength": 1, "lora_8_strength": 1, "toggle": false, "lora_1_clip_strength": 1, "lora_2_name": "None", "lora_10_name": "None", "lora_10_clip_strength": 1, "lora_10_strength": 1, "lora_1_model_strength": 1, "lora_5_name": "None", "lora_3_model_strength": 1, "lora_9_strength": 1, "lora_3_strength": 1, "lora_5_clip_strength": 1, "lora_6_model_strength": 1, "lora_6_strength": 1, "lora_7_clip_strength": 1, "lora_9_name": "None" }, "_meta": { "title": "简易Lora堆" } }, "85": { "class_type": "LayerUtility: SaveImagePlus", "inputs": { "preview": false, "filename_prefix": "comfyui", "images": [ "47", 5 ], "save_workflow_as_json": false, "custom_path": "", "format": "png", "meta_data": false, "blind_watermark": "", "timestamp": "None", "quality": 100 }, "_meta": { "title": "LayerUtility: SaveImage Plus" } }, "86": { "class_type": "CropFace", "inputs": { "image": [ "16", 0 ], "facedetection": "retinaface_resnet50" }, "_meta": { "title": "裁剪面部" } }, "43": { "class_type": "ApplyInstantID", "inputs": { "instantid": [ "44", 0 ], "end_at": 1, "image": [ "86", 0 ], "negative": [ "11", 2 ], "control_net": [ "46", 0 ], "insightface": [ "45", 0 ], "image_kps": [ "89", 0 ], "weight": 0.8, "model": [ "11", 0 ], "positive": [ "11", 1 ], "start_at": 0 }, "_meta": { "title": "应用InstantID" } }, "87": { "class_type": "ImageUpscaleWithModel", "inputs": { "image": [ "11", 5 ], "upscale_model": [ "88", 0 ] }, "_meta": { "title": "图像通过模型放大" } } }

Logs

No response

Other

有哪位大佬帮忙解答一下? 目前的现象就是使用同样的prompt ,手动使用web页面就可以成功在Comfyui/output目录下生成图片 ,而使用websockets_api_example.py和websockets_api_example_wx_images.py中的代码请求comfyui服务 返回结果就是空的 代码没动 只换了prompt 下面就是我的prompt

{ "22": { "class_type": "PulidModelLoader", "inputs": { "pulid_file": "ip-adapter_pulid_sdxl_fp16.safetensors" }, "_meta": { "title": "PuLID模型加载器" } }, "44": { "class_type": "InstantIDModelLoader", "inputs": { "instantid_file": "ip-adapter.bin" }, "_meta": { "title": "InstnatID模型加载器" } }, "88": { "class_type": "UpscaleModelLoader", "inputs": { "model_name": "4x_NMKD-Siax_200k.pth" }, "_meta": { "title": "放大模型加载器" } }, "23": { "class_type": "PulidEvaClipLoader", "inputs": {}, "_meta": { "title": "PuLIDEVAClip加载器" } }, "45": { "class_type": "InstantIDFaceAnalysis", "inputs": { "provider": "CPU" }, "_meta": { "title": "InstantID面部分析" } }, "89": { "class_type": "ImageScaleBy", "inputs": { "image": [ "87", 0 ], "scale_by": 0.38, "upscale_method": "nearest-exact" }, "_meta": { "title": "图像按系数缩放" } }, "46": { "class_type": "ControlNetLoader", "inputs": { "control_net_name": "diffusion_pytorch_model.safetensors" }, "_meta": { "title": "ControlNet加载器" } }, "47": { "class_type": "KSampler (Efficient)", "inputs": { "vae_decode": "true", "latent_image": [ "90", 0 ], "seed": [ "93", 0 ], "cfg": 5, "positive": [ "43", 1 ], "steps": 25, "script": [ "75", 0 ], "scheduler": "karras", "negative": [ "43", 2 ], "denoise": 0.58, "sampler_name": "dpmpp_2m_sde", "optional_vae": [ "11", 4 ], "preview_method": "none", "model": [ "43", 0 ] }, "_meta": { "title": "K采样器(效率)" } }, "29": { "class_type": "ApplyPulid", "inputs": { "end_at": 1, "image": [ "86", 0 ], "pulid": [ "22", 0 ], "method": "fidelity", "weight": 0.3, "model": [ "10", 0 ], "eva_clip": [ "23", 0 ], "face_analysis": [ "76", 0 ], "start_at": 0 }, "_meta": { "title": "应用PuLID" } }, "90": { "class_type": "VAEEncode", "inputs": { "pixels": [ "89", 0 ], "vae": [ "11", 4 ] }, "_meta": { "title": "VAE编码" } }, "92": { "class_type": "String Literal", "inputs": { "string": "professional,4k,highly detailed,medium portrait soft light, beautiful model,vivid,photorealistic, vivid colors, ", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "93": { "class_type": "Seed Generator", "inputs": { "seed": 1047523952660796 }, "_meta": { "title": "Seed Generator" } }, "94": { "class_type": "String Literal", "inputs": { "string": "1man,indoors,business suit,necktie,,upper_body,portrait, ", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "75": { "class_type": "Noise Control Script", "inputs": { "seed": 813662192183951, "cfg_denoiser": false, "add_seed_noise": false, "weight": 0.015, "rng_source": "gpu" }, "_meta": { "title": "控制噪波" } }, "10": { "class_type": "Efficient Loader", "inputs": { "lora_name": "None", "weight_interpretation": "A1111", "batch_size": 2, "empty_latent_width": 832, "empty_latent_height": 1216, "speak_and_recognation": true, "token_normalization": "mean", "lora_stack": [ "103", 0 ], "positive": [ "99", 0 ], "lora_clip_strength": 1, "ckpt_name": "leosamsHelloworldXL_helloworldXL70.safetensors", "negative": "(worst quality, low quality, blurry, bad eye, wrong hand, bad anatomy, wrong anatomy, open mouth, deformed, distorted, disfigured, cgi, illustration, cartoon, poorly drawn, watermark),nsfw,nipples,flag,american_flag, ", "lora_model_strength": 1, "clip_skip": -1, "vae_name": "Baked VAE" }, "_meta": { "title": "效率加载器" } }, "76": { "class_type": "PulidInsightFaceLoader", "inputs": { "provider": "CPU" }, "_meta": { "title": "PuLIDInsightFace加载器" } }, "11": { "class_type": "KSampler (Efficient)", "inputs": { "vae_decode": "true", "latent_image": [ "10", 3 ], "seed": [ "93", 0 ], "cfg": 6, "positive": [ "10", 1 ], "steps": 25, "script": [ "75", 0 ], "scheduler": "karras", "negative": [ "10", 2 ], "denoise": 1, "sampler_name": "dpmpp_2m_sde", "optional_vae": [ "10", 4 ], "preview_method": "none", "model": [ "29", 0 ] }, "_meta": { "title": "K采样器(效率)" } }, "99": { "class_type": "Text Concatenate", "inputs": { "delimiter": ", ", "text_a": [ "94", 0 ], "clean_whitespace": "true", "text_c": [ "92", 0 ], "text_b": [ "100", 0 ] }, "_meta": { "title": "文本连锁" } }, "16": { "class_type": "LoadImage", "inputs": { "image": "94a506e555b24c20969b19a401825986.png", "upload": "image" }, "_meta": { "title": "加载图像" } }, "100": { "class_type": "String Literal", "inputs": { "string": "", "speak_and_recognation": true }, "_meta": { "title": "String Literal" } }, "103": { "class_type": "easy loraStack", "inputs": { "lora_3_name": "None", "lora_6_name": "None", "lora_7_strength": 1, "lora_9_clip_strength": 1, "lora_1_strength": 0.6, "lora_9_model_strength": 1, "lora_5_model_strength": 1, "lora_4_clip_strength": 1, "lora_2_clip_strength": 1, "lora_4_strength": 1, "mode": "simple", "num_loras": 1, "lora_6_clip_strength": 1, "lora_8_name": "None", "lora_7_name": "None", "lora_8_model_strength": 1, "lora_5_strength": 1, "lora_4_name": "None", "lora_4_model_strength": 1, "lora_1_name": "None", "lora_2_strength": 1, "lora_2_model_strength": 1, "lora_3_clip_strength": 1, "lora_8_clip_strength": 1, "lora_10_model_strength": 1, "lora_7_model_strength": 1, "lora_8_strength": 1, "toggle": false, "lora_1_clip_strength": 1, "lora_2_name": "None", "lora_10_name": "None", "lora_10_clip_strength": 1, "lora_10_strength": 1, "lora_1_model_strength": 1, "lora_5_name": "None", "lora_3_model_strength": 1, "lora_9_strength": 1, "lora_3_strength": 1, "lora_5_clip_strength": 1, "lora_6_model_strength": 1, "lora_6_strength": 1, "lora_7_clip_strength": 1, "lora_9_name": "None" }, "_meta": { "title": "简易Lora堆" } }, "85": { "class_type": "LayerUtility: SaveImagePlus", "inputs": { "preview": false, "filename_prefix": "comfyui", "images": [ "47", 5 ], "save_workflow_as_json": false, "custom_path": "", "format": "png", "meta_data": false, "blind_watermark": "", "timestamp": "None", "quality": 100 }, "_meta": { "title": "LayerUtility: SaveImage Plus" } }, "86": { "class_type": "CropFace", "inputs": { "image": [ "16", 0 ], "facedetection": "retinaface_resnet50" }, "_meta": { "title": "裁剪面部" } }, "43": { "class_type": "ApplyInstantID", "inputs": { "instantid": [ "44", 0 ], "end_at": 1, "image": [ "86", 0 ], "negative": [ "11", 2 ], "control_net": [ "46", 0 ], "insightface": [ "45", 0 ], "image_kps": [ "89", 0 ], "weight": 0.8, "model": [ "11", 0 ], "positive": [ "11", 1 ], "start_at": 0 }, "_meta": { "title": "应用InstantID" } }, "87": { "class_type": "ImageUpscaleWithModel", "inputs": { "image": [ "11", 5 ], "upscale_model": [ "88", 0 ] }, "_meta": { "title": "图像通过模型放大" } } }

github-actions[bot] commented 1 month ago

This issue is being marked stale because it has not had any activity for 30 days. Reply below within 7 days if your issue still isn't solved, and it will be left open. Otherwise, the issue will be closed automatically.