Files
SnapOtter/docker/feature-manifest.json
T
SnapOtterandGitHub 44348e6e07 fix(ai): constrain tokenizers and onnxruntime, drop obsolete arm64 protobuf pins (#693)
The worst-order install test caught tokenizers 0.23.1 against transformers' <0.21 requirement, and the r2 rebuilds drifted onnxruntime. Both are constrained now; the obsolete protobuf<5 arm64 pins are gone (arm64 resolves mediapipe 1.0.0, no python protobuf dependency); transcription and object-eraser-colorize point at r3 archives. Full worst-order install plus the 18-tool installed AI matrix passed against this manifest. Completes #669.
2026-07-30 21:43:28 +08:00

588 lines
22 KiB
JSON

{
"manifestVersion": 2,
"imageVersion": "2.0.0",
"pythonVersion": {
"amd64": "3.12",
"arm64": "3.11"
},
"basePackages": [
"numpy==1.26.4",
"Pillow==12.3.0",
"opencv-python-headless==4.10.0.84",
"huggingface-hub[hf_xet,hf_transfer]==0.36.2"
],
"constraints": [
"numpy==1.26.4",
"scipy==1.12.0",
"scikit-learn==1.4.2",
"scikit-image==0.24.0",
"pandas==2.2.2",
"huggingface-hub==0.36.2",
"tokenizers==0.20.3",
"onnxruntime==1.20.1"
],
"bundleRepo": "deepsafe/feature-bundles",
"bundles": {
"background-removal": {
"name": "Background Removal",
"description": "Remove image backgrounds with AI",
"estimatedSize": "4-5 GB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/background-removal-amd64-gpu.tar.gz",
"sha256": "a1dfbdd5862fd32c7719ca79b0c7e4b76a327330a5991ddad98a2761abca10cb",
"compressedSize": 4808242472,
"extractedSize": 0
},
"arm64-cpu": {
"file": "v2.0.0/background-removal-arm64-cpu-r2.tar.gz",
"sha256": "eb453211098ca6a90b23bae987c3be6873cf7f0f7d67e1855d34f335a2add076",
"compressedSize": 4535687017,
"extractedSize": 6375426284
}
},
"packages": {
"common": ["rembg==2.0.62"],
"amd64": ["onnxruntime-gpu==1.20.1", "mediapipe>=0.10.21"],
"arm64": ["onnxruntime==1.20.1", "mediapipe>=0.10.18"]
},
"pipFlags": {},
"postInstall": [],
"models": [
{
"id": "rembg-u2net",
"downloadFn": "rembg_session",
"args": ["u2net"],
"sha256": "8d10d2f3bb75ae3b6d527c77944fc5e7dcd94b29809d47a739a7a728a912b491"
},
{
"id": "rembg-isnet-general-use",
"downloadFn": "rembg_session",
"args": ["isnet-general-use"],
"sha256": "60920e99c45464f2ba57bee2ad08c919a52bbf852739e96947fbb4358c0d964a"
},
{
"id": "rembg-bria-rmbg",
"downloadFn": "rembg_session",
"args": ["bria-rmbg"],
"sha256": "5b486f08200f513f460da46dd701db5fbb47d79b4be4b708a19444bcd4e79958"
},
{
"id": "rembg-birefnet-general-lite",
"downloadFn": "rembg_session",
"args": ["birefnet-general-lite"],
"sha256": "5600024376f572a557870a5eb0afb1e5961636bef4e1e22132025467d0f03333"
},
{
"id": "rembg-birefnet-portrait",
"downloadFn": "rembg_session",
"args": ["birefnet-portrait"],
"sha256": "1ba1c8ff5a7bbfadc8d8d13fb11d7be793f91f23d9d466549e37a854f6668f99"
},
{
"id": "rembg-birefnet-general",
"downloadFn": "rembg_session",
"args": ["birefnet-general"],
"sha256": "58f621f00f5d756097615970a88a791584600dcf7c45b18a0a6267535a1ebd3c"
},
{
"id": "rembg-birefnet-matting",
"downloadFn": "rembg_session",
"args": ["birefnet-matting"],
"sha256": "6065d27c615ea27308f5b88598dd8db116eb07436c7a323ca40d13b2866c309e"
},
{
"id": "rembg-birefnet-hr-matting",
"downloadFn": "rembg_session",
"args": ["birefnet-hr-matting"],
"sha256": "45b7d92ce0e2e75e1c8a564a9e84c87ac9da7f4315d2deeef3674864e809aef3"
}
],
"smokeImports": ["rembg", "onnxruntime"],
"enablesTools": [
"remove-background",
"remove-gif-background",
"passport-photo",
"transparency-fixer",
"background-replace",
"blur-background"
]
},
"face-detection": {
"name": "Face Detection",
"description": "Detect and blur faces, fix red-eye, smart crop",
"estimatedSize": "200-300 MB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/face-detection-amd64-gpu.tar.gz",
"sha256": "0ee28b86dac88c60e005604ba0595746f455ec97f08d0c8307ba9b57e131bc7f",
"compressedSize": 75358868,
"extractedSize": 209838080
},
"arm64-cpu": {
"file": "v2.0.0/face-detection-arm64-cpu-r2.tar.gz",
"sha256": "c66d494c03beb3988decb2eea528ba4dced95956f337855ee6efee9b95f03720",
"compressedSize": 81361023,
"extractedSize": 232238876
}
},
"packages": {
"common": [],
"amd64": ["mediapipe>=0.10.21"],
"arm64": ["mediapipe>=0.10.18"]
},
"pipFlags": {},
"postInstall": [],
"models": [
{
"id": "mediapipe-face-detector",
"url": "https://storage.googleapis.com/mediapipe-models/face_detector/blaze_face_short_range/float16/latest/blaze_face_short_range.tflite",
"sha256": "b4578f35940bf5a1a655214a1cce5cab13eba73c1297cd78e1a04c2380b0152f",
"path": "mediapipe/blaze_face_short_range.tflite",
"minSize": 100000
},
{
"id": "mediapipe-face-landmarker",
"url": "https://storage.googleapis.com/mediapipe-models/face_landmarker/face_landmarker/float16/latest/face_landmarker.task",
"sha256": "64184e229b263107bc2b804c6625db1341ff2bb731874b0bcc2fe6544e0bc9ff",
"path": "mediapipe/face_landmarker.task",
"minSize": 1000000
}
],
"smokeImports": ["mediapipe"],
"enablesTools": ["blur-faces", "red-eye-removal", "smart-crop"]
},
"object-eraser-colorize": {
"name": "Object Eraser & Colorize",
"description": "Erase objects from photos and colorize B&W images",
"estimatedSize": "1-2 GB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/object-eraser-colorize-amd64-gpu.tar.gz",
"sha256": "a3f6cd2eca9e5dd907497d0d58ca5f6532887bb748e4062de0f4fb9815782c31",
"compressedSize": 1545472964,
"extractedSize": 2249379840
},
"arm64-cpu": {
"file": "v2.0.0/object-eraser-colorize-arm64-cpu-r3.tar.gz",
"sha256": "18aaf0bc507e94e248296f3092e625ff6339365b6df7030439c4ec9465098bfa",
"compressedSize": 1253914457,
"extractedSize": 1437568952
}
},
"packages": {
"common": ["huggingface-hub[hf_xet]==0.36.2"],
"amd64": ["onnxruntime-gpu==1.20.1"],
"arm64": ["onnxruntime==1.20.1"]
},
"pipFlags": {},
"postInstall": [],
"models": [
{
"id": "lama-onnx",
"downloadFn": "hf_snapshot",
"args": ["Carve/LaMa-ONNX", "lama"],
"revision": "c3c0c9e468934d62e79c329e35d82dd09ff8c444",
"file": "lama_fp32.onnx",
"path": "lama/lama_fp32.onnx",
"minSize": 100000000
},
{
"id": "ddcolor-onnx",
"downloadFn": "hf_snapshot",
"args": ["facefusion/models-3.0.0", "ddcolor"],
"revision": "728b9659bd9691bf32cbf7f61af478d94b7ba81e",
"file": "ddcolor.onnx",
"path": "ddcolor/ddcolor.onnx",
"minSize": 50000000
},
{
"id": "opencv-colorize-prototxt",
"url": "https://raw.githubusercontent.com/richzhang/colorization/caffe/colorization/models/colorization_deploy_v2.prototxt",
"sha256": "d16418cef8df4ccd703a55ae0ef3960861d5010418f77c90d0a47689998a7169",
"path": "colorize-opencv/colorization_deploy_v2.prototxt",
"minSize": 0
},
{
"id": "opencv-colorize-caffemodel",
"downloadFn": "hf_snapshot",
"args": ["BilalSardar/Black-N-White-To-Color", "colorize-opencv"],
"revision": "41c04215906bbd3d8e2a09c07b3e76f42f0cd65e",
"file": "colorization_release_v2.caffemodel",
"path": "colorize-opencv/colorization_release_v2.caffemodel",
"minSize": 100000000,
"repoType": "space"
},
{
"id": "opencv-colorize-points",
"url": "https://raw.githubusercontent.com/richzhang/colorization/caffe/colorization/resources/pts_in_hull.npy",
"sha256": "b5dec01315c34f43f1c8c089e84c45ae35d1838d8e77ed0e7ca930f79ffa450e",
"path": "colorize-opencv/pts_in_hull.npy",
"minSize": 0
}
],
"smokeImports": ["onnxruntime"],
"enablesTools": ["erase-object", "colorize", "ai-canvas-expand"]
},
"inpaint-hq": {
"name": "High-Quality Inpainting",
"description": "Diffusion-based object removal for large objects and detailed textures",
"estimatedSize": "5-7 GB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/inpaint-hq-amd64-gpu.tar.gz",
"sha256": "da65c2b7e678b50815c7c2d8ae4594ad7bbf4c1e84682bfc91b46948cdff71de",
"compressedSize": 7283076233,
"extractedSize": 10745029298
},
"arm64-cpu": {
"file": "v2.0.0/inpaint-hq-arm64-cpu.tar.gz",
"sha256": "b854b101a6190dde94d9d0039129941122c62df7c9a6429d425231cf4fa4630f",
"compressedSize": 5130985609,
"extractedSize": 7159491153
}
},
"packages": {
"common": [
"diffusers==0.31.0",
"transformers==4.46.3",
"accelerate==1.2.1",
"safetensors==0.4.5",
"huggingface-hub[hf_xet]==0.36.2"
],
"amd64": [
"torch==2.7.0+cu126 torchvision==0.22.0+cu126 --index-url https://download.pytorch.org/whl/cu126"
],
"arm64": [
"torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cpu"
]
},
"pipFlags": {},
"postInstall": [],
"models": [
{
"id": "sd15-inpainting",
"downloadFn": "hf_snapshot",
"args": ["stable-diffusion-v1-5/stable-diffusion-inpainting", "sd15-inpainting"],
"revision": "8a4288a76071f7280aedbdb3253bdb9e9d5d84bb",
"allowPatterns": ["*.json", "*.txt", "tokenizer/*", "*.fp16.safetensors"],
"path": "sd15-inpainting/unet/diffusion_pytorch_model.fp16.safetensors",
"minSize": 1000000000
}
],
"smokeImports": ["diffusers", "torch", "transformers"],
"enablesTools": ["erase-object"]
},
"upscale-enhance": {
"name": "Upscale & Enhance",
"description": "AI upscaling, face enhancement, and noise removal",
"estimatedSize": "4-5 GB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/upscale-enhance-amd64-gpu.tar.gz",
"sha256": "b725a0b18a8295cb87e15f79b31c294f1de98101dedd76389728300a8d377a15",
"compressedSize": 5358573124,
"extractedSize": 8406943960
},
"arm64-cpu": {
"file": "v2.0.0/upscale-enhance-arm64-cpu-r2.tar.gz",
"sha256": "d2d9745d2ca1b546c4e5399d5bf8c024bad349135fdb1ad0e4da28b7ac19fc9e",
"compressedSize": 2209931629,
"extractedSize": 3246159800
}
},
"packages": {
"common": [
"codeformer-pip==0.0.4",
"huggingface-hub[hf_xet]==0.36.2",
"einops",
"setuptools<75"
],
"amd64": [
"torch==2.7.0+cu126 torchvision==0.22.0+cu126 --index-url https://download.pytorch.org/whl/cu126",
"lpips",
"basicsr==1.4.2",
"realesrgan==0.3.0",
"mediapipe>=0.10.21"
],
"arm64": [
"torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cpu",
"lpips",
"basicsr==1.4.2",
"realesrgan==0.3.0",
"mediapipe>=0.10.18"
]
},
"pipFlags": {
"codeformer-pip==0.0.4": "--no-deps",
"basicsr==1.4.2": "--no-build-isolation"
},
"postInstall": ["numpy==1.26.4"],
"models": [
{
"id": "realesrgan-x4plus",
"url": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
"sha256": "4fa0d38905f75ac06eb49a7951b426670021be3018265fd191d2125df9d682f1",
"path": "realesrgan/RealESRGAN_x4plus.pth",
"minSize": 60000000
},
{
"id": "realesrgan-x2plus",
"url": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth",
"sha256": "49fafd45f8fd7aa8d31ab2a22d14d91b536c34494a5cfe31eb5d89c2fa266abb",
"path": "realesrgan/RealESRGAN_x2plus.pth",
"minSize": 60000000
},
{
"id": "gfpgan-v1.3",
"url": "https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.3.pth",
"sha256": "c953a88f2727c85c3d9ae72e2bd4846bbaf59fe6972ad94130e23e7017524a70",
"path": "gfpgan/GFPGANv1.3.pth",
"minSize": 300000000
},
{
"id": "codeformer-pth",
"url": "https://github.com/sczhou/CodeFormer/releases/download/v0.1.0/codeformer.pth",
"sha256": "1009e537e0c2a07d4cabce6355f53cb66767cd4b4297ec7a4a64ca4b8a5684b7",
"path": "codeformer/codeformer.pth",
"minSize": 350000000
},
{
"id": "codeformer-onnx",
"downloadFn": "hf_snapshot",
"args": ["facefusion/models-3.0.0", "codeformer"],
"revision": "728b9659bd9691bf32cbf7f61af478d94b7ba81e",
"file": "codeformer.onnx",
"path": "codeformer/codeformer.onnx",
"minSize": 100000000
},
{
"id": "facexlib-detection",
"url": "https://github.com/xinntao/facexlib/releases/download/v0.1.0/detection_Resnet50_Final.pth",
"sha256": "6d1de9c2944f2ccddca5f5e010ea5ae64a39845a86311af6fdf30841b0a5a16d",
"path": "gfpgan/facelib/detection_Resnet50_Final.pth",
"minSize": 100000000
},
{
"id": "facexlib-parsing",
"url": "https://github.com/xinntao/facexlib/releases/download/v0.2.2/parsing_parsenet.pth",
"sha256": "3d558d8d0e42c20224f13cf5a29c79eba2d59913419f945545d8cf7b72920de2",
"path": "gfpgan/facelib/parsing_parsenet.pth",
"minSize": 80000000
},
{
"id": "scunet-color-real",
"url": "https://github.com/cszn/KAIR/releases/download/v1.0/scunet_color_real_psnr.pth",
"sha256": "fa78899ba2caec9d235a900e91d96c689da71c42029230c2028b00f09f809c2e",
"path": "scunet/scunet_color_real_psnr.pth",
"minSize": 3000000
},
{
"id": "nafnet-sidd",
"downloadFn": "hf_snapshot",
"args": ["mikestealth/nafnet-models", "nafnet"],
"revision": "cd3a91ddc41c9c3a53ee809b3ddd5263c9e94e51",
"file": "NAFNet-SIDD-width64.pth",
"path": "nafnet/NAFNet-SIDD-width64.pth",
"minSize": 60000000
}
],
"patches": [
{
"file": "basicsr/data/degradations.py",
"search": "from torchvision.transforms.functional_tensor import rgb_to_grayscale",
"replace": "from torchvision.transforms.functional import rgb_to_grayscale"
}
],
"smokeImports": ["torch", "torchvision"],
"enablesTools": ["upscale", "enhance-faces", "noise-removal"]
},
"photo-restoration": {
"name": "Photo Restoration",
"description": "Restore old or damaged photos",
"estimatedSize": "800 MB - 1 GB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/photo-restoration-amd64-gpu.tar.gz",
"sha256": "4747953db35c71b83db19ce12cc6c591730f0cb9f64621f239db17ae3e3563fd",
"compressedSize": 4680254837,
"extractedSize": 8172949791
},
"arm64-cpu": {
"file": "v2.0.0/photo-restoration-arm64-cpu-r2.tar.gz",
"sha256": "86f1b261314f9f66146c02fddcbeff1880da0b52716e21126c75c7eee00e445d",
"compressedSize": 1252233816,
"extractedSize": 2238681871
}
},
"packages": {
"common": ["codeformer-pip==0.0.4", "huggingface-hub[hf_xet]==0.36.2", "setuptools<75"],
"amd64": [
"onnxruntime-gpu==1.20.1",
"mediapipe>=0.10.21",
"torch==2.7.0+cu126 torchvision==0.22.0+cu126 --index-url https://download.pytorch.org/whl/cu126",
"lpips",
"basicsr==1.4.2",
"realesrgan==0.3.0"
],
"arm64": [
"torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cpu",
"onnxruntime==1.20.1",
"mediapipe>=0.10.18",
"lpips",
"basicsr==1.4.2",
"realesrgan==0.3.0"
]
},
"pipFlags": {
"codeformer-pip==0.0.4": "--no-deps",
"basicsr==1.4.2": "--no-build-isolation"
},
"postInstall": ["numpy==1.26.4"],
"models": [
{
"id": "lama-onnx",
"downloadFn": "hf_snapshot",
"args": ["Carve/LaMa-ONNX", "lama"],
"revision": "c3c0c9e468934d62e79c329e35d82dd09ff8c444",
"file": "lama_fp32.onnx",
"path": "lama/lama_fp32.onnx",
"minSize": 100000000
},
{
"id": "codeformer-onnx",
"downloadFn": "hf_snapshot",
"args": ["facefusion/models-3.0.0", "codeformer"],
"revision": "728b9659bd9691bf32cbf7f61af478d94b7ba81e",
"file": "codeformer.onnx",
"path": "codeformer/codeformer.onnx",
"minSize": 100000000
},
{
"id": "realesrgan-x4plus",
"url": "https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth",
"sha256": "4fa0d38905f75ac06eb49a7951b426670021be3018265fd191d2125df9d682f1",
"path": "realesrgan/RealESRGAN_x4plus.pth",
"minSize": 60000000
},
{
"id": "facexlib-detection",
"url": "https://github.com/xinntao/facexlib/releases/download/v0.1.0/detection_Resnet50_Final.pth",
"sha256": "6d1de9c2944f2ccddca5f5e010ea5ae64a39845a86311af6fdf30841b0a5a16d",
"path": "gfpgan/facelib/detection_Resnet50_Final.pth",
"minSize": 100000000
},
{
"id": "facexlib-parsing",
"url": "https://github.com/xinntao/facexlib/releases/download/v0.2.2/parsing_parsenet.pth",
"sha256": "3d558d8d0e42c20224f13cf5a29c79eba2d59913419f945545d8cf7b72920de2",
"path": "gfpgan/facelib/parsing_parsenet.pth",
"minSize": 80000000
},
{
"id": "mediapipe-face-detector",
"url": "https://storage.googleapis.com/mediapipe-models/face_detector/blaze_face_short_range/float16/latest/blaze_face_short_range.tflite",
"sha256": "b4578f35940bf5a1a655214a1cce5cab13eba73c1297cd78e1a04c2380b0152f",
"path": "mediapipe/blaze_face_short_range.tflite",
"minSize": 100000
},
{
"id": "mediapipe-face-landmarker",
"url": "https://storage.googleapis.com/mediapipe-models/face_landmarker/face_landmarker/float16/latest/face_landmarker.task",
"sha256": "64184e229b263107bc2b804c6625db1341ff2bb731874b0bcc2fe6544e0bc9ff",
"path": "mediapipe/face_landmarker.task",
"minSize": 1000000
},
{
"id": "scunet-color-real",
"url": "https://github.com/cszn/KAIR/releases/download/v1.0/scunet_color_real_psnr.pth",
"sha256": "fa78899ba2caec9d235a900e91d96c689da71c42029230c2028b00f09f809c2e",
"path": "scunet/scunet_color_real_psnr.pth",
"minSize": 3000000
}
],
"patches": [
{
"file": "basicsr/data/degradations.py",
"search": "from torchvision.transforms.functional_tensor import rgb_to_grayscale",
"replace": "from torchvision.transforms.functional import rgb_to_grayscale"
}
],
"smokeImports": ["onnxruntime", "torch", "mediapipe"],
"enablesTools": ["restore-photo"]
},
"ocr": {
"name": "OCR",
"description": "High-accuracy local OCR for images and PDFs",
"estimatedSize": "~208-234 MiB download / ~409-488 MiB installed",
"runtimeFormatVersion": 3,
"runtimeFamily": "ocr",
"modelManifest": "docker/ocr-runtime-models.json",
"targets": {
"linux-amd64-cpu-py312": {
"file": "v3/ocr-linux-amd64-cpu-py312.tar.gz",
"compressedSizeEstimate": 244622462,
"extractedSizeEstimate": 510762511,
"sizeKind": "measured-estimate",
"minimumMemoryBytes": 4294967296,
"requirements": "docker/ocr-runtime-requirements-amd64.txt"
},
"linux-arm64-cpu-py311": {
"file": "v3/ocr-linux-arm64-cpu-py311.tar.gz",
"compressedSizeEstimate": 218031562,
"extractedSizeEstimate": 428565655,
"sizeKind": "measured-estimate",
"minimumMemoryBytes": 4294967296,
"requirements": "docker/ocr-runtime-requirements-arm64.txt"
}
},
"packages": {
"common": ["rapidocr==3.9.1", "onnxruntime==1.20.1"],
"amd64": [],
"arm64": []
},
"pipFlags": {},
"postInstall": [],
"models": [],
"smokeImports": ["rapidocr", "onnxruntime"],
"enablesTools": ["ocr", "ocr-pdf"]
},
"transcription": {
"name": "Transcription",
"description": "Speech to text for audio and video (subtitles)",
"estimatedSize": "~600 MB",
"archives": {
"amd64-gpu": {
"file": "v2.0.0/transcription-amd64-gpu.tar.gz",
"sha256": "3033d480c6183df66855d78c355233dbc86a754d07414e544e1df1038337c9e1",
"compressedSize": 559189061,
"extractedSize": 845987840
},
"arm64-cpu": {
"file": "v2.0.0/transcription-arm64-cpu-r3.tar.gz",
"sha256": "19938e1220bdcfdbdb0e6c5019971b38b4a519f818d9cd29b789bc19de661485",
"compressedSize": 536230905,
"extractedSize": 781074121
}
},
"packages": {
"common": ["faster-whisper>=1.0.0", "huggingface-hub[hf_xet]==0.36.2"],
"amd64": [],
"arm64": []
},
"pipFlags": {},
"postInstall": [],
"models": [
{
"id": "faster-whisper-small",
"downloadFn": "hf_snapshot",
"args": ["Systran/faster-whisper-small", "faster-whisper-small"],
"revision": "536b0662742c02347bc0e980a01041f333bce120"
}
],
"smokeImports": ["faster_whisper", "ctranslate2"],
"enablesTools": ["transcribe-audio", "auto-subtitles"]
}
}
}