Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Self-Flow-XL-2-256/demo.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,95 @@
+---
+library_name: diffusers
+pipeline_tag: unconditional-image-generation
+tags:
+  - diffusers
+  - self-flow
+  - image-generation
+  - class-conditional
+  - imagenet
+  - flow-matching
+license: apache-2.0
+inference: true
+widget:
+  - output:
+      url: Self-Flow-XL-2-256/demo.png
+language:
+  - en
+---
+
+# Self-Flow-diffusers
+
+Diffusers-ready checkpoints for **Self-Flow** (Self-Supervised Flow Matching), converted from the [official Self-Flow release](https://github.com/black-forest-labs/Self-Flow) for local/offline use.
+
+This root folder is a model collection that contains:
+
+- `Self-Flow-XL-2-256`
+
+Each subfolder is a self-contained Diffusers model repo with:
+
+- `pipeline.py` (`SelfFlowPipeline`)
+- `transformer/transformer_selfflow.py` and weights
+- `scheduler/scheduling_flow_match_selfflow.py` (`SelfFlowFlowMatchScheduler`, SDE flow matching)
+- `scheduler/scheduler_config.json`
+- `vae/` (`stabilityai/sd-vae-ft-ema`)
+
+Each variant embeds English `id2label` in `model_index.json`, so class labels can be passed as ImageNet ids or English synonym strings.
+
+## Demo
+
+![Self-Flow-XL-2-256 demo](Self-Flow-XL-2-256/demo.png)
+
+Class-conditional sample (ImageNet class **207**, golden retriever), `Self-Flow-XL/2` at 256×256, 250 steps, CFG 3.5, seed 42.
+
+## Model Paths
+
+Use paths relative to this root README:
+
+| Model | Resolution | Local path |
+| --- | ---: | --- |
+| Self-Flow-XL/2 | 256×256 | `./Self-Flow-XL-2-256` |
+
+## Recommended inference
+
+| Setting | Value |
+| --- | --- |
+| Resolution | 256×256 |
+| Sampler | Self-Flow SDE flow matching |
+| Steps | 250 |
+| CFG scale | 3.5 |
+| Guidance interval | `(0.0, 0.7)` when CFG > 1 |
+| Dtype | `bfloat16` (recommended on Ampere+) |
+| VAE | `stabilityai/sd-vae-ft-ema` |
+
+## Inference Demo (Diffusers)
+
+```python
+from pathlib import Path
+import torch
+from diffusers import DiffusionPipeline
+
+model_dir = Path("./Self-Flow-XL-2-256").resolve()
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+)
+pipe.to(device)
+
+generator = torch.Generator(device=device).manual_seed(42)
+
+print(pipe.id2label[207])
+print(pipe.get_label_ids("golden retriever"))  # [207]
+
+image = pipe(
+    class_labels="golden retriever",
+    num_inference_steps=250,
+    guidance_scale=3.5,
+    generator=generator,
+).images[0]
+image.save("self_flow_xl_256_demo.png")
+```

Self-Flow-XL-2-256/README.md

@@ -0,0 +1,63 @@
+---
+license: apache-2.0
+library_name: diffusers
+pipeline_tag: unconditional-image-generation
+tags:
+  - diffusers
+  - self-flow
+  - image-generation
+  - class-conditional
+  - imagenet
+inference: true
+---
+
+# Self-Flow-XL-2-256
+
+Self-contained Diffusers checkpoint for **Self-Flow-XL/2** on ImageNet 256×256, converted from the official Self-Flow release.
+
+## Recommended inference
+
+| Setting | Value |
+| --- | --- |
+| Resolution | 256×256 |
+| Sampler | Self-Flow SDE flow matching |
+| Steps | 250 |
+| CFG scale | 3.5 |
+| Guidance interval | `(0.0, 0.7)` when CFG > 1 |
+| Dtype | `bfloat16` (recommended on Ampere+) |
+| VAE | `stabilityai/sd-vae-ft-ema` |
+
+```python
+from pathlib import Path
+import torch
+from diffusers import DiffusionPipeline
+
+model_dir = Path("./Self-Flow-XL-2-256").resolve()
+pipe = DiffusionPipeline.from_pretrained(
+    str(model_dir),
+    local_files_only=True,
+    custom_pipeline=str(model_dir / "pipeline.py"),
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+)
+pipe.to("cuda")
+
+generator = torch.Generator(device="cuda").manual_seed(42)
+image = pipe(
+    class_labels="golden retriever",
+    num_inference_steps=250,
+    guidance_scale=3.5,
+    generator=generator,
+).images[0]
+image.save("demo.png")
+```
+
+## Components
+
+- `pipeline.py`
+- `token_utils.py`
+- `transformer/transformer_selfflow.py`
+- `transformer/diffusion_pytorch_model.safetensors`
+- `scheduler/scheduling_flow_match_selfflow.py`
+- `scheduler/scheduler_config.json`
+- `vae_pretrained_model_name_or_path.txt`

Self-Flow-XL-2-256/model_index.json

@@ -0,0 +1,1021 @@
+{
+  "_class_name": [
+    "pipeline",
+    "SelfFlowPipeline"
+  ],
+  "_diffusers_version": "0.36.0",
+  "transformer": [
+    "transformer_selfflow",
+    "SelfFlowTransformer2DModel"
+  ],
+  "scheduler": [
+    "scheduling_flow_match_selfflow",
+    "SelfFlowFlowMatchScheduler"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ],
+  "id2label": {
+    "0": "tench, Tinca tinca",
+    "1": "goldfish, Carassius auratus",
+    "2": "great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias",
+    "3": "tiger shark, Galeocerdo cuvieri",
+    "4": "hammerhead, hammerhead shark",
+    "5": "electric ray, crampfish, numbfish, torpedo",
+    "6": "stingray",
+    "7": "cock",
+    "8": "hen",
+    "9": "ostrich, Struthio camelus",
+    "10": "brambling, Fringilla montifringilla",
+    "11": "goldfinch, Carduelis carduelis",
+    "12": "house finch, linnet, Carpodacus mexicanus",
+    "13": "junco, snowbird",
+    "14": "indigo bunting, indigo finch, indigo bird, Passerina cyanea",
+    "15": "robin, American robin, Turdus migratorius",
+    "16": "bulbul",
+    "17": "jay",
+    "18": "magpie",
+    "19": "chickadee",
+    "20": "water ouzel, dipper",
+    "21": "kite",
+    "22": "bald eagle, American eagle, Haliaeetus leucocephalus",
+    "23": "vulture",
+    "24": "great grey owl, great gray owl, Strix nebulosa",
+    "25": "European fire salamander, Salamandra salamandra",
+    "26": "common newt, Triturus vulgaris",
+    "27": "eft",
+    "28": "spotted salamander, Ambystoma maculatum",
+    "29": "axolotl, mud puppy, Ambystoma mexicanum",
+    "30": "bullfrog, Rana catesbeiana",
+    "31": "tree frog, tree-frog",
+    "32": "tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui",
+    "33": "loggerhead, loggerhead turtle, Caretta caretta",
+    "34": "leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea",
+    "35": "mud turtle",
+    "36": "terrapin",
+    "37": "box turtle, box tortoise",
+    "38": "banded gecko",
+    "39": "common iguana, iguana, Iguana iguana",
+    "40": "American chameleon, anole, Anolis carolinensis",
+    "41": "whiptail, whiptail lizard",
+    "42": "agama",
+    "43": "frilled lizard, Chlamydosaurus kingi",
+    "44": "alligator lizard",
+    "45": "Gila monster, Heloderma suspectum",
+    "46": "green lizard, Lacerta viridis",
+    "47": "African chameleon, Chamaeleo chamaeleon",
+    "48": "Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis",
+    "49": "African crocodile, Nile crocodile, Crocodylus niloticus",
+    "50": "American alligator, Alligator mississipiensis",
+    "51": "triceratops",
+    "52": "thunder snake, worm snake, Carphophis amoenus",
+    "53": "ringneck snake, ring-necked snake, ring snake",
+    "54": "hognose snake, puff adder, sand viper",
+    "55": "green snake, grass snake",
+    "56": "king snake, kingsnake",
+    "57": "garter snake, grass snake",
+    "58": "water snake",
+    "59": "vine snake",
+    "60": "night snake, Hypsiglena torquata",
+    "61": "boa constrictor, Constrictor constrictor",
+    "62": "rock python, rock snake, Python sebae",
+    "63": "Indian cobra, Naja naja",
+    "64": "green mamba",
+    "65": "sea snake",
+    "66": "horned viper, cerastes, sand viper, horned asp, Cerastes cornutus",
+    "67": "diamondback, diamondback rattlesnake, Crotalus adamanteus",
+    "68": "sidewinder, horned rattlesnake, Crotalus cerastes",
+    "69": "trilobite",
+    "70": "harvestman, daddy longlegs, Phalangium opilio",
+    "71": "scorpion",
+    "72": "black and gold garden spider, Argiope aurantia",
+    "73": "barn spider, Araneus cavaticus",
+    "74": "garden spider, Aranea diademata",
+    "75": "black widow, Latrodectus mactans",
+    "76": "tarantula",
+    "77": "wolf spider, hunting spider",
+    "78": "tick",
+    "79": "centipede",
+    "80": "black grouse",
+    "81": "ptarmigan",
+    "82": "ruffed grouse, partridge, Bonasa umbellus",
+    "83": "prairie chicken, prairie grouse, prairie fowl",
+    "84": "peacock",
+    "85": "quail",
+    "86": "partridge",
+    "87": "African grey, African gray, Psittacus erithacus",
+    "88": "macaw",
+    "89": "sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita",
+    "90": "lorikeet",
+    "91": "coucal",
+    "92": "bee eater",
+    "93": "hornbill",
+    "94": "hummingbird",
+    "95": "jacamar",
+    "96": "toucan",
+    "97": "drake",
+    "98": "red-breasted merganser, Mergus serrator",
+    "99": "goose",
+    "100": "black swan, Cygnus atratus",
+    "101": "tusker",
+    "102": "echidna, spiny anteater, anteater",
+    "103": "platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus",
+    "104": "wallaby, brush kangaroo",
+    "105": "koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus",
+    "106": "wombat",
+    "107": "jellyfish",
+    "108": "sea anemone, anemone",
+    "109": "brain coral",
+    "110": "flatworm, platyhelminth",
+    "111": "nematode, nematode worm, roundworm",
+    "112": "conch",
+    "113": "snail",
+    "114": "slug",
+    "115": "sea slug, nudibranch",
+    "116": "chiton, coat-of-mail shell, sea cradle, polyplacophore",
+    "117": "chambered nautilus, pearly nautilus, nautilus",
+    "118": "Dungeness crab, Cancer magister",
+    "119": "rock crab, Cancer irroratus",
+    "120": "fiddler crab",
+    "121": "king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica",
+    "122": "American lobster, Northern lobster, Maine lobster, Homarus americanus",
+    "123": "spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish",
+    "124": "crayfish, crawfish, crawdad, crawdaddy",
+    "125": "hermit crab",
+    "126": "isopod",
+    "127": "white stork, Ciconia ciconia",
+    "128": "black stork, Ciconia nigra",
+    "129": "spoonbill",
+    "130": "flamingo",
+    "131": "little blue heron, Egretta caerulea",
+    "132": "American egret, great white heron, Egretta albus",
+    "133": "bittern",
+    "134": "crane",
+    "135": "limpkin, Aramus pictus",
+    "136": "European gallinule, Porphyrio porphyrio",
+    "137": "American coot, marsh hen, mud hen, water hen, Fulica americana",
+    "138": "bustard",
+    "139": "ruddy turnstone, Arenaria interpres",
+    "140": "red-backed sandpiper, dunlin, Erolia alpina",
+    "141": "redshank, Tringa totanus",
+    "142": "dowitcher",
+    "143": "oystercatcher, oyster catcher",
+    "144": "pelican",
+    "145": "king penguin, Aptenodytes patagonica",
+    "146": "albatross, mollymawk",
+    "147": "grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus",
+    "148": "killer whale, killer, orca, grampus, sea wolf, Orcinus orca",
+    "149": "dugong, Dugong dugon",
+    "150": "sea lion",
+    "151": "Chihuahua",
+    "152": "Japanese spaniel",
+    "153": "Maltese dog, Maltese terrier, Maltese",
+    "154": "Pekinese, Pekingese, Peke",
+    "155": "Shih-Tzu",
+    "156": "Blenheim spaniel",
+    "157": "papillon",
+    "158": "toy terrier",
+    "159": "Rhodesian ridgeback",
+    "160": "Afghan hound, Afghan",
+    "161": "basset, basset hound",
+    "162": "beagle",
+    "163": "bloodhound, sleuthhound",
+    "164": "bluetick",
+    "165": "black-and-tan coonhound",
+    "166": "Walker hound, Walker foxhound",
+    "167": "English foxhound",
+    "168": "redbone",
+    "169": "borzoi, Russian wolfhound",
+    "170": "Irish wolfhound",
+    "171": "Italian greyhound",
+    "172": "whippet",
+    "173": "Ibizan hound, Ibizan Podenco",
+    "174": "Norwegian elkhound, elkhound",
+    "175": "otterhound, otter hound",
+    "176": "Saluki, gazelle hound",
+    "177": "Scottish deerhound, deerhound",
+    "178": "Weimaraner",
+    "179": "Staffordshire bullterrier, Staffordshire bull terrier",
+    "180": "American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier",
+    "181": "Bedlington terrier",
+    "182": "Border terrier",
+    "183": "Kerry blue terrier",
+    "184": "Irish terrier",
+    "185": "Norfolk terrier",
+    "186": "Norwich terrier",
+    "187": "Yorkshire terrier",
+    "188": "wire-haired fox terrier",
+    "189": "Lakeland terrier",
+    "190": "Sealyham terrier, Sealyham",
+    "191": "Airedale, Airedale terrier",
+    "192": "cairn, cairn terrier",
+    "193": "Australian terrier",
+    "194": "Dandie Dinmont, Dandie Dinmont terrier",
+    "195": "Boston bull, Boston terrier",
+    "196": "miniature schnauzer",
+    "197": "giant schnauzer",
+    "198": "standard schnauzer",
+    "199": "Scotch terrier, Scottish terrier, Scottie",
+    "200": "Tibetan terrier, chrysanthemum dog",
+    "201": "silky terrier, Sydney silky",
+    "202": "soft-coated wheaten terrier",
+    "203": "West Highland white terrier",
+    "204": "Lhasa, Lhasa apso",
+    "205": "flat-coated retriever",
+    "206": "curly-coated retriever",
+    "207": "golden retriever",
+    "208": "Labrador retriever",
+    "209": "Chesapeake Bay retriever",
+    "210": "German short-haired pointer",
+    "211": "vizsla, Hungarian pointer",
+    "212": "English setter",
+    "213": "Irish setter, red setter",
+    "214": "Gordon setter",
+    "215": "Brittany spaniel",
+    "216": "clumber, clumber spaniel",
+    "217": "English springer, English springer spaniel",
+    "218": "Welsh springer spaniel",
+    "219": "cocker spaniel, English cocker spaniel, cocker",
+    "220": "Sussex spaniel",
+    "221": "Irish water spaniel",
+    "222": "kuvasz",
+    "223": "schipperke",
+    "224": "groenendael",
+    "225": "malinois",
+    "226": "briard",
+    "227": "kelpie",
+    "228": "komondor",
+    "229": "Old English sheepdog, bobtail",
+    "230": "Shetland sheepdog, Shetland sheep dog, Shetland",
+    "231": "collie",
+    "232": "Border collie",
+    "233": "Bouvier des Flandres, Bouviers des Flandres",
+    "234": "Rottweiler",
+    "235": "German shepherd, German shepherd dog, German police dog, alsatian",
+    "236": "Doberman, Doberman pinscher",
+    "237": "miniature pinscher",
+    "238": "Greater Swiss Mountain dog",
+    "239": "Bernese mountain dog",
+    "240": "Appenzeller",
+    "241": "EntleBucher",
+    "242": "boxer",
+    "243": "bull mastiff",
+    "244": "Tibetan mastiff",
+    "245": "French bulldog",
+    "246": "Great Dane",
+    "247": "Saint Bernard, St Bernard",
+    "248": "Eskimo dog, husky",
+    "249": "malamute, malemute, Alaskan malamute",
+    "250": "Siberian husky",
+    "251": "dalmatian, coach dog, carriage dog",
+    "252": "affenpinscher, monkey pinscher, monkey dog",
+    "253": "basenji",
+    "254": "pug, pug-dog",
+    "255": "Leonberg",
+    "256": "Newfoundland, Newfoundland dog",
+    "257": "Great Pyrenees",
+    "258": "Samoyed, Samoyede",
+    "259": "Pomeranian",
+    "260": "chow, chow chow",
+    "261": "keeshond",
+    "262": "Brabancon griffon",
+    "263": "Pembroke, Pembroke Welsh corgi",
+    "264": "Cardigan, Cardigan Welsh corgi",
+    "265": "toy poodle",
+    "266": "miniature poodle",
+    "267": "standard poodle",
+    "268": "Mexican hairless",
+    "269": "timber wolf, grey wolf, gray wolf, Canis lupus",
+    "270": "white wolf, Arctic wolf, Canis lupus tundrarum",
+    "271": "red wolf, maned wolf, Canis rufus, Canis niger",
+    "272": "coyote, prairie wolf, brush wolf, Canis latrans",
+    "273": "dingo, warrigal, warragal, Canis dingo",
+    "274": "dhole, Cuon alpinus",
+    "275": "African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus",
+    "276": "hyena, hyaena",
+    "277": "red fox, Vulpes vulpes",
+    "278": "kit fox, Vulpes macrotis",
+    "279": "Arctic fox, white fox, Alopex lagopus",
+    "280": "grey fox, gray fox, Urocyon cinereoargenteus",
+    "281": "tabby, tabby cat",
+    "282": "tiger cat",
+    "283": "Persian cat",
+    "284": "Siamese cat, Siamese",
+    "285": "Egyptian cat",
+    "286": "cougar, puma, catamount, mountain lion, painter, panther, Felis concolor",
+    "287": "lynx, catamount",
+    "288": "leopard, Panthera pardus",
+    "289": "snow leopard, ounce, Panthera uncia",
+    "290": "jaguar, panther, Panthera onca, Felis onca",
+    "291": "lion, king of beasts, Panthera leo",
+    "292": "tiger, Panthera tigris",
+    "293": "cheetah, chetah, Acinonyx jubatus",
+    "294": "brown bear, bruin, Ursus arctos",
+    "295": "American black bear, black bear, Ursus americanus, Euarctos americanus",
+    "296": "ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus",
+    "297": "sloth bear, Melursus ursinus, Ursus ursinus",
+    "298": "mongoose",
+    "299": "meerkat, mierkat",
+    "300": "tiger beetle",
+    "301": "ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle",
+    "302": "ground beetle, carabid beetle",
+    "303": "long-horned beetle, longicorn, longicorn beetle",
+    "304": "leaf beetle, chrysomelid",
+    "305": "dung beetle",
+    "306": "rhinoceros beetle",
+    "307": "weevil",
+    "308": "fly",
+    "309": "bee",
+    "310": "ant, emmet, pismire",
+    "311": "grasshopper, hopper",
+    "312": "cricket",
+    "313": "walking stick, walkingstick, stick insect",
+    "314": "cockroach, roach",
+    "315": "mantis, mantid",
+    "316": "cicada, cicala",
+    "317": "leafhopper",
+    "318": "lacewing, lacewing fly",
+    "319": "dragonfly, darning needle, devils darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
+    "320": "damselfly",
+    "321": "admiral",
+    "322": "ringlet, ringlet butterfly",
+    "323": "monarch, monarch butterfly, milkweed butterfly, Danaus plexippus",
+    "324": "cabbage butterfly",
+    "325": "sulphur butterfly, sulfur butterfly",
+    "326": "lycaenid, lycaenid butterfly",
+    "327": "starfish, sea star",
+    "328": "sea urchin",
+    "329": "sea cucumber, holothurian",
+    "330": "wood rabbit, cottontail, cottontail rabbit",
+    "331": "hare",
+    "332": "Angora, Angora rabbit",
+    "333": "hamster",
+    "334": "porcupine, hedgehog",
+    "335": "fox squirrel, eastern fox squirrel, Sciurus niger",
+    "336": "marmot",
+    "337": "beaver",
+    "338": "guinea pig, Cavia cobaya",
+    "339": "sorrel",
+    "340": "zebra",
+    "341": "hog, pig, grunter, squealer, Sus scrofa",
+    "342": "wild boar, boar, Sus scrofa",
+    "343": "warthog",
+    "344": "hippopotamus, hippo, river horse, Hippopotamus amphibius",
+    "345": "ox",
+    "346": "water buffalo, water ox, Asiatic buffalo, Bubalus bubalis",
+    "347": "bison",
+    "348": "ram, tup",
+    "349": "bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis",
+    "350": "ibex, Capra ibex",
+    "351": "hartebeest",
+    "352": "impala, Aepyceros melampus",
+    "353": "gazelle",
+    "354": "Arabian camel, dromedary, Camelus dromedarius",
+    "355": "llama",
+    "356": "weasel",
+    "357": "mink",
+    "358": "polecat, fitch, foulmart, foumart, Mustela putorius",
+    "359": "black-footed ferret, ferret, Mustela nigripes",
+    "360": "otter",
+    "361": "skunk, polecat, wood pussy",
+    "362": "badger",
+    "363": "armadillo",
+    "364": "three-toed sloth, ai, Bradypus tridactylus",
+    "365": "orangutan, orang, orangutang, Pongo pygmaeus",
+    "366": "gorilla, Gorilla gorilla",
+    "367": "chimpanzee, chimp, Pan troglodytes",
+    "368": "gibbon, Hylobates lar",
+    "369": "siamang, Hylobates syndactylus, Symphalangus syndactylus",
+    "370": "guenon, guenon monkey",
+    "371": "patas, hussar monkey, Erythrocebus patas",
+    "372": "baboon",
+    "373": "macaque",
+    "374": "langur",
+    "375": "colobus, colobus monkey",
+    "376": "proboscis monkey, Nasalis larvatus",
+    "377": "marmoset",
+    "378": "capuchin, ringtail, Cebus capucinus",
+    "379": "howler monkey, howler",
+    "380": "titi, titi monkey",
+    "381": "spider monkey, Ateles geoffroyi",
+    "382": "squirrel monkey, Saimiri sciureus",
+    "383": "Madagascar cat, ring-tailed lemur, Lemur catta",
+    "384": "indri, indris, Indri indri, Indri brevicaudatus",
+    "385": "Indian elephant, Elephas maximus",
+    "386": "African elephant, Loxodonta africana",
+    "387": "lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens",
+    "388": "giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca",
+    "389": "barracouta, snoek",
+    "390": "eel",
+    "391": "coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch",
+    "392": "rock beauty, Holocanthus tricolor",
+    "393": "anemone fish",
+    "394": "sturgeon",
+    "395": "gar, garfish, garpike, billfish, Lepisosteus osseus",
+    "396": "lionfish",
+    "397": "puffer, pufferfish, blowfish, globefish",
+    "398": "abacus",
+    "399": "abaya",
+    "400": "academic gown, academic robe, judge robe",
+    "401": "accordion, piano accordion, squeeze box",
+    "402": "acoustic guitar",
+    "403": "aircraft carrier, carrier, flattop, attack aircraft carrier",
+    "404": "airliner",
+    "405": "airship, dirigible",
+    "406": "altar",
+    "407": "ambulance",
+    "408": "amphibian, amphibious vehicle",
+    "409": "analog clock",
+    "410": "apiary, bee house",
+    "411": "apron",
+    "412": "ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin",
+    "413": "assault rifle, assault gun",
+    "414": "backpack, back pack, knapsack, packsack, rucksack, haversack",
+    "415": "bakery, bakeshop, bakehouse",
+    "416": "balance beam, beam",
+    "417": "balloon",
+    "418": "ballpoint, ballpoint pen, ballpen, Biro",
+    "419": "Band Aid",
+    "420": "banjo",
+    "421": "bannister, banister, balustrade, balusters, handrail",
+    "422": "barbell",
+    "423": "barber chair",
+    "424": "barbershop",
+    "425": "barn",
+    "426": "barometer",
+    "427": "barrel, cask",
+    "428": "barrow, garden cart, lawn cart, wheelbarrow",
+    "429": "baseball",
+    "430": "basketball",
+    "431": "bassinet",
+    "432": "bassoon",
+    "433": "bathing cap, swimming cap",
+    "434": "bath towel",
+    "435": "bathtub, bathing tub, bath, tub",
+    "436": "beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon",
+    "437": "beacon, lighthouse, beacon light, pharos",
+    "438": "beaker",
+    "439": "bearskin, busby, shako",
+    "440": "beer bottle",
+    "441": "beer glass",
+    "442": "bell cote, bell cot",
+    "443": "bib",
+    "444": "bicycle-built-for-two, tandem bicycle, tandem",
+    "445": "bikini, two-piece",
+    "446": "binder, ring-binder",
+    "447": "binoculars, field glasses, opera glasses",
+    "448": "birdhouse",
+    "449": "boathouse",
+    "450": "bobsled, bobsleigh, bob",
+    "451": "bolo tie, bolo, bola tie, bola",
+    "452": "bonnet, poke bonnet",
+    "453": "bookcase",
+    "454": "bookshop, bookstore, bookstall",
+    "455": "bottlecap",
+    "456": "bow",
+    "457": "bow tie, bow-tie, bowtie",
+    "458": "brass, memorial tablet, plaque",
+    "459": "brassiere, bra, bandeau",
+    "460": "breakwater, groin, groyne, mole, bulwark, seawall, jetty",
+    "461": "breastplate, aegis, egis",
+    "462": "broom",
+    "463": "bucket, pail",
+    "464": "buckle",
+    "465": "bulletproof vest",
+    "466": "bullet train, bullet",
+    "467": "butcher shop, meat market",
+    "468": "cab, hack, taxi, taxicab",
+    "469": "caldron, cauldron",
+    "470": "candle, taper, wax light",
+    "471": "cannon",
+    "472": "canoe",
+    "473": "can opener, tin opener",
+    "474": "cardigan",
+    "475": "car mirror",
+    "476": "carousel, carrousel, merry-go-round, roundabout, whirligig",
+    "477": "carpenters kit, tool kit",
+    "478": "carton",
+    "479": "car wheel",
+    "480": "cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM",
+    "481": "cassette",
+    "482": "cassette player",
+    "483": "castle",
+    "484": "catamaran",
+    "485": "CD player",
+    "486": "cello, violoncello",
+    "487": "cellular telephone, cellular phone, cellphone, cell, mobile phone",
+    "488": "chain",
+    "489": "chainlink fence",
+    "490": "chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour",
+    "491": "chain saw, chainsaw",
+    "492": "chest",
+    "493": "chiffonier, commode",
+    "494": "chime, bell, gong",
+    "495": "china cabinet, china closet",
+    "496": "Christmas stocking",
+    "497": "church, church building",
+    "498": "cinema, movie theater, movie theatre, movie house, picture palace",
+    "499": "cleaver, meat cleaver, chopper",
+    "500": "cliff dwelling",
+    "501": "cloak",
+    "502": "clog, geta, patten, sabot",
+    "503": "cocktail shaker",
+    "504": "coffee mug",
+    "505": "coffeepot",
+    "506": "coil, spiral, volute, whorl, helix",
+    "507": "combination lock",
+    "508": "computer keyboard, keypad",
+    "509": "confectionery, confectionary, candy store",
+    "510": "container ship, containership, container vessel",
+    "511": "convertible",
+    "512": "corkscrew, bottle screw",
+    "513": "cornet, horn, trumpet, trump",
+    "514": "cowboy boot",
+    "515": "cowboy hat, ten-gallon hat",
+    "516": "cradle",
+    "517": "crane",
+    "518": "crash helmet",
+    "519": "crate",
+    "520": "crib, cot",
+    "521": "Crock Pot",
+    "522": "croquet ball",
+    "523": "crutch",
+    "524": "cuirass",
+    "525": "dam, dike, dyke",
+    "526": "desk",
+    "527": "desktop computer",
+    "528": "dial telephone, dial phone",
+    "529": "diaper, nappy, napkin",
+    "530": "digital clock",
+    "531": "digital watch",
+    "532": "dining table, board",
+    "533": "dishrag, dishcloth",
+    "534": "dishwasher, dish washer, dishwashing machine",
+    "535": "disk brake, disc brake",
+    "536": "dock, dockage, docking facility",
+    "537": "dogsled, dog sled, dog sleigh",
+    "538": "dome",
+    "539": "doormat, welcome mat",
+    "540": "drilling platform, offshore rig",
+    "541": "drum, membranophone, tympan",
+    "542": "drumstick",
+    "543": "dumbbell",
+    "544": "Dutch oven",
+    "545": "electric fan, blower",
+    "546": "electric guitar",
+    "547": "electric locomotive",
+    "548": "entertainment center",
+    "549": "envelope",
+    "550": "espresso maker",
+    "551": "face powder",
+    "552": "feather boa, boa",
+    "553": "file, file cabinet, filing cabinet",
+    "554": "fireboat",
+    "555": "fire engine, fire truck",
+    "556": "fire screen, fireguard",
+    "557": "flagpole, flagstaff",
+    "558": "flute, transverse flute",
+    "559": "folding chair",
+    "560": "football helmet",
+    "561": "forklift",
+    "562": "fountain",
+    "563": "fountain pen",
+    "564": "four-poster",
+    "565": "freight car",
+    "566": "French horn, horn",
+    "567": "frying pan, frypan, skillet",
+    "568": "fur coat",
+    "569": "garbage truck, dustcart",
+    "570": "gasmask, respirator, gas helmet",
+    "571": "gas pump, gasoline pump, petrol pump, island dispenser",
+    "572": "goblet",
+    "573": "go-kart",
+    "574": "golf ball",
+    "575": "golfcart, golf cart",
+    "576": "gondola",
+    "577": "gong, tam-tam",
+    "578": "gown",
+    "579": "grand piano, grand",
+    "580": "greenhouse, nursery, glasshouse",
+    "581": "grille, radiator grille",
+    "582": "grocery store, grocery, food market, market",
+    "583": "guillotine",
+    "584": "hair slide",
+    "585": "hair spray",
+    "586": "half track",
+    "587": "hammer",
+    "588": "hamper",
+    "589": "hand blower, blow dryer, blow drier, hair dryer, hair drier",
+    "590": "hand-held computer, hand-held microcomputer",
+    "591": "handkerchief, hankie, hanky, hankey",
+    "592": "hard disc, hard disk, fixed disk",
+    "593": "harmonica, mouth organ, harp, mouth harp",
+    "594": "harp",
+    "595": "harvester, reaper",
+    "596": "hatchet",
+    "597": "holster",
+    "598": "home theater, home theatre",
+    "599": "honeycomb",
+    "600": "hook, claw",
+    "601": "hoopskirt, crinoline",
+    "602": "horizontal bar, high bar",
+    "603": "horse cart, horse-cart",
+    "604": "hourglass",
+    "605": "iPod",
+    "606": "iron, smoothing iron",
+    "607": "jack-o-lantern",
+    "608": "jean, blue jean, denim",
+    "609": "jeep, landrover",
+    "610": "jersey, T-shirt, tee shirt",
+    "611": "jigsaw puzzle",
+    "612": "jinrikisha, ricksha, rickshaw",
+    "613": "joystick",
+    "614": "kimono",
+    "615": "knee pad",
+    "616": "knot",
+    "617": "lab coat, laboratory coat",
+    "618": "ladle",
+    "619": "lampshade, lamp shade",
+    "620": "laptop, laptop computer",
+    "621": "lawn mower, mower",
+    "622": "lens cap, lens cover",
+    "623": "letter opener, paper knife, paperknife",
+    "624": "library",
+    "625": "lifeboat",
+    "626": "lighter, light, igniter, ignitor",
+    "627": "limousine, limo",
+    "628": "liner, ocean liner",
+    "629": "lipstick, lip rouge",
+    "630": "Loafer",
+    "631": "lotion",
+    "632": "loudspeaker, speaker, speaker unit, loudspeaker system, speaker system",
+    "633": "loupe, jewelers loupe",
+    "634": "lumbermill, sawmill",
+    "635": "magnetic compass",
+    "636": "mailbag, postbag",
+    "637": "mailbox, letter box",
+    "638": "maillot",
+    "639": "maillot, tank suit",
+    "640": "manhole cover",
+    "641": "maraca",
+    "642": "marimba, xylophone",
+    "643": "mask",
+    "644": "matchstick",
+    "645": "maypole",
+    "646": "maze, labyrinth",
+    "647": "measuring cup",
+    "648": "medicine chest, medicine cabinet",
+    "649": "megalith, megalithic structure",
+    "650": "microphone, mike",
+    "651": "microwave, microwave oven",
+    "652": "military uniform",
+    "653": "milk can",
+    "654": "minibus",
+    "655": "miniskirt, mini",
+    "656": "minivan",
+    "657": "missile",
+    "658": "mitten",
+    "659": "mixing bowl",
+    "660": "mobile home, manufactured home",
+    "661": "Model T",
+    "662": "modem",
+    "663": "monastery",
+    "664": "monitor",
+    "665": "moped",
+    "666": "mortar",
+    "667": "mortarboard",
+    "668": "mosque",
+    "669": "mosquito net",
+    "670": "motor scooter, scooter",
+    "671": "mountain bike, all-terrain bike, off-roader",
+    "672": "mountain tent",
+    "673": "mouse, computer mouse",
+    "674": "mousetrap",
+    "675": "moving van",
+    "676": "muzzle",
+    "677": "nail",
+    "678": "neck brace",
+    "679": "necklace",
+    "680": "nipple",
+    "681": "notebook, notebook computer",
+    "682": "obelisk",
+    "683": "oboe, hautboy, hautbois",
+    "684": "ocarina, sweet potato",
+    "685": "odometer, hodometer, mileometer, milometer",
+    "686": "oil filter",
+    "687": "organ, pipe organ",
+    "688": "oscilloscope, scope, cathode-ray oscilloscope, CRO",
+    "689": "overskirt",
+    "690": "oxcart",
+    "691": "oxygen mask",
+    "692": "packet",
+    "693": "paddle, boat paddle",
+    "694": "paddlewheel, paddle wheel",
+    "695": "padlock",
+    "696": "paintbrush",
+    "697": "pajama, pyjama, pjs, jammies",
+    "698": "palace",
+    "699": "panpipe, pandean pipe, syrinx",
+    "700": "paper towel",
+    "701": "parachute, chute",
+    "702": "parallel bars, bars",
+    "703": "park bench",
+    "704": "parking meter",
+    "705": "passenger car, coach, carriage",
+    "706": "patio, terrace",
+    "707": "pay-phone, pay-station",
+    "708": "pedestal, plinth, footstall",
+    "709": "pencil box, pencil case",
+    "710": "pencil sharpener",
+    "711": "perfume, essence",
+    "712": "Petri dish",
+    "713": "photocopier",
+    "714": "pick, plectrum, plectron",
+    "715": "pickelhaube",
+    "716": "picket fence, paling",
+    "717": "pickup, pickup truck",
+    "718": "pier",
+    "719": "piggy bank, penny bank",
+    "720": "pill bottle",
+    "721": "pillow",
+    "722": "ping-pong ball",
+    "723": "pinwheel",
+    "724": "pirate, pirate ship",
+    "725": "pitcher, ewer",
+    "726": "plane, carpenters plane, woodworking plane",
+    "727": "planetarium",
+    "728": "plastic bag",
+    "729": "plate rack",
+    "730": "plow, plough",
+    "731": "plunger, plumbers helper",
+    "732": "Polaroid camera, Polaroid Land camera",
+    "733": "pole",
+    "734": "police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria",
+    "735": "poncho",
+    "736": "pool table, billiard table, snooker table",
+    "737": "pop bottle, soda bottle",
+    "738": "pot, flowerpot",
+    "739": "potters wheel",
+    "740": "power drill",
+    "741": "prayer rug, prayer mat",
+    "742": "printer",
+    "743": "prison, prison house",
+    "744": "projectile, missile",
+    "745": "projector",
+    "746": "puck, hockey puck",
+    "747": "punching bag, punch bag, punching ball, punchball",
+    "748": "purse",
+    "749": "quill, quill pen",
+    "750": "quilt, comforter, comfort, puff",
+    "751": "racer, race car, racing car",
+    "752": "racket, racquet",
+    "753": "radiator",
+    "754": "radio, wireless",
+    "755": "radio telescope, radio reflector",
+    "756": "rain barrel",
+    "757": "recreational vehicle, RV, R.V.",
+    "758": "reel",
+    "759": "reflex camera",
+    "760": "refrigerator, icebox",
+    "761": "remote control, remote",
+    "762": "restaurant, eating house, eating place, eatery",
+    "763": "revolver, six-gun, six-shooter",
+    "764": "rifle",
+    "765": "rocking chair, rocker",
+    "766": "rotisserie",
+    "767": "rubber eraser, rubber, pencil eraser",
+    "768": "rugby ball",
+    "769": "rule, ruler",
+    "770": "running shoe",
+    "771": "safe",
+    "772": "safety pin",
+    "773": "saltshaker, salt shaker",
+    "774": "sandal",
+    "775": "sarong",
+    "776": "sax, saxophone",
+    "777": "scabbard",
+    "778": "scale, weighing machine",
+    "779": "school bus",
+    "780": "schooner",
+    "781": "scoreboard",
+    "782": "screen, CRT screen",
+    "783": "screw",
+    "784": "screwdriver",
+    "785": "seat belt, seatbelt",
+    "786": "sewing machine",
+    "787": "shield, buckler",
+    "788": "shoe shop, shoe-shop, shoe store",
+    "789": "shoji",
+    "790": "shopping basket",
+    "791": "shopping cart",
+    "792": "shovel",
+    "793": "shower cap",
+    "794": "shower curtain",
+    "795": "ski",
+    "796": "ski mask",
+    "797": "sleeping bag",
+    "798": "slide rule, slipstick",
+    "799": "sliding door",
+    "800": "slot, one-armed bandit",
+    "801": "snorkel",
+    "802": "snowmobile",
+    "803": "snowplow, snowplough",
+    "804": "soap dispenser",
+    "805": "soccer ball",
+    "806": "sock",
+    "807": "solar dish, solar collector, solar furnace",
+    "808": "sombrero",
+    "809": "soup bowl",
+    "810": "space bar",
+    "811": "space heater",
+    "812": "space shuttle",
+    "813": "spatula",
+    "814": "speedboat",
+    "815": "spider web, spiders web",
+    "816": "spindle",
+    "817": "sports car, sport car",
+    "818": "spotlight, spot",
+    "819": "stage",
+    "820": "steam locomotive",
+    "821": "steel arch bridge",
+    "822": "steel drum",
+    "823": "stethoscope",
+    "824": "stole",
+    "825": "stone wall",
+    "826": "stopwatch, stop watch",
+    "827": "stove",
+    "828": "strainer",
+    "829": "streetcar, tram, tramcar, trolley, trolley car",
+    "830": "stretcher",
+    "831": "studio couch, day bed",
+    "832": "stupa, tope",
+    "833": "submarine, pigboat, sub, U-boat",
+    "834": "suit, suit of clothes",
+    "835": "sundial",
+    "836": "sunglass",
+    "837": "sunglasses, dark glasses, shades",
+    "838": "sunscreen, sunblock, sun blocker",
+    "839": "suspension bridge",
+    "840": "swab, swob, mop",
+    "841": "sweatshirt",
+    "842": "swimming trunks, bathing trunks",
+    "843": "swing",
+    "844": "switch, electric switch, electrical switch",
+    "845": "syringe",
+    "846": "table lamp",
+    "847": "tank, army tank, armored combat vehicle, armoured combat vehicle",
+    "848": "tape player",
+    "849": "teapot",
+    "850": "teddy, teddy bear",
+    "851": "television, television system",
+    "852": "tennis ball",
+    "853": "thatch, thatched roof",
+    "854": "theater curtain, theatre curtain",
+    "855": "thimble",
+    "856": "thresher, thrasher, threshing machine",
+    "857": "throne",
+    "858": "tile roof",
+    "859": "toaster",
+    "860": "tobacco shop, tobacconist shop, tobacconist",
+    "861": "toilet seat",
+    "862": "torch",
+    "863": "totem pole",
+    "864": "tow truck, tow car, wrecker",
+    "865": "toyshop",
+    "866": "tractor",
+    "867": "trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi",
+    "868": "tray",
+    "869": "trench coat",
+    "870": "tricycle, trike, velocipede",
+    "871": "trimaran",
+    "872": "tripod",
+    "873": "triumphal arch",
+    "874": "trolleybus, trolley coach, trackless trolley",
+    "875": "trombone",
+    "876": "tub, vat",
+    "877": "turnstile",
+    "878": "typewriter keyboard",
+    "879": "umbrella",
+    "880": "unicycle, monocycle",
+    "881": "upright, upright piano",
+    "882": "vacuum, vacuum cleaner",
+    "883": "vase",
+    "884": "vault",
+    "885": "velvet",
+    "886": "vending machine",
+    "887": "vestment",
+    "888": "viaduct",
+    "889": "violin, fiddle",
+    "890": "volleyball",
+    "891": "waffle iron",
+    "892": "wall clock",
+    "893": "wallet, billfold, notecase, pocketbook",
+    "894": "wardrobe, closet, press",
+    "895": "warplane, military plane",
+    "896": "washbasin, handbasin, washbowl, lavabo, wash-hand basin",
+    "897": "washer, automatic washer, washing machine",
+    "898": "water bottle",
+    "899": "water jug",
+    "900": "water tower",
+    "901": "whiskey jug",
+    "902": "whistle",
+    "903": "wig",
+    "904": "window screen",
+    "905": "window shade",
+    "906": "Windsor tie",
+    "907": "wine bottle",
+    "908": "wing",
+    "909": "wok",
+    "910": "wooden spoon",
+    "911": "wool, woolen, woollen",
+    "912": "worm fence, snake fence, snake-rail fence, Virginia fence",
+    "913": "wreck",
+    "914": "yawl",
+    "915": "yurt",
+    "916": "web site, website, internet site, site",
+    "917": "comic book",
+    "918": "crossword puzzle, crossword",
+    "919": "street sign",
+    "920": "traffic light, traffic signal, stoplight",
+    "921": "book jacket, dust cover, dust jacket, dust wrapper",
+    "922": "menu",
+    "923": "plate",
+    "924": "guacamole",
+    "925": "consomme",
+    "926": "hot pot, hotpot",
+    "927": "trifle",
+    "928": "ice cream, icecream",
+    "929": "ice lolly, lolly, lollipop, popsicle",
+    "930": "French loaf",
+    "931": "bagel, beigel",
+    "932": "pretzel",
+    "933": "cheeseburger",
+    "934": "hotdog, hot dog, red hot",
+    "935": "mashed potato",
+    "936": "head cabbage",
+    "937": "broccoli",
+    "938": "cauliflower",
+    "939": "zucchini, courgette",
+    "940": "spaghetti squash",
+    "941": "acorn squash",
+    "942": "butternut squash",
+    "943": "cucumber, cuke",
+    "944": "artichoke, globe artichoke",
+    "945": "bell pepper",
+    "946": "cardoon",
+    "947": "mushroom",
+    "948": "Granny Smith",
+    "949": "strawberry",
+    "950": "orange",
+    "951": "lemon",
+    "952": "fig",
+    "953": "pineapple, ananas",
+    "954": "banana",
+    "955": "jackfruit, jak, jack",
+    "956": "custard apple",
+    "957": "pomegranate",
+    "958": "hay",
+    "959": "carbonara",
+    "960": "chocolate sauce, chocolate syrup",
+    "961": "dough",
+    "962": "meat loaf, meatloaf",
+    "963": "pizza, pizza pie",
+    "964": "potpie",
+    "965": "burrito",
+    "966": "red wine",
+    "967": "espresso",
+    "968": "cup",
+    "969": "eggnog",
+    "970": "alp",
+    "971": "bubble",
+    "972": "cliff, drop, drop-off",
+    "973": "coral reef",
+    "974": "geyser",
+    "975": "lakeside, lakeshore",
+    "976": "promontory, headland, head, foreland",
+    "977": "sandbar, sand bar",
+    "978": "seashore, coast, seacoast, sea-coast",
+    "979": "valley, vale",
+    "980": "volcano",
+    "981": "ballplayer, baseball player",
+    "982": "groom, bridegroom",
+    "983": "scuba diver",
+    "984": "rapeseed",
+    "985": "daisy",
+    "986": "yellow ladys slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
+    "987": "corn",
+    "988": "acorn",
+    "989": "hip, rose hip, rosehip",
+    "990": "buckeye, horse chestnut, conker",
+    "991": "coral fungus",
+    "992": "agaric",
+    "993": "gyromitra",
+    "994": "stinkhorn, carrion fungus",
+    "995": "earthstar",
+    "996": "hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa",
+    "997": "bolete",
+    "998": "ear, spike, capitulum",
+    "999": "toilet tissue, toilet paper, bathroom tissue"
+  }
+}

Self-Flow-XL-2-256/pipeline.py

@@ -0,0 +1,477 @@
+"""Hub custom pipeline: SelfFlowPipeline.
+Load with native Hugging Face diffusers and trust_remote_code=True.
+"""
+
+from __future__ import annotations
+
+import importlib
+import inspect
+import json
+import sys
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+import torch
+from einops import rearrange
+
+from diffusers import AutoencoderKL, FlowMatchEulerDiscreteScheduler
+from diffusers.image_processor import VaeImageProcessor
+from diffusers.pipelines.pipeline_utils import DiffusionPipeline, ImagePipelineOutput
+from diffusers.schedulers import KarrasDiffusionSchedulers
+from diffusers.utils.torch_utils import randn_tensor
+
+DEFAULT_LATENT_SIZE = 32
+DEFAULT_IMAGE_SIZE = 256
+
+
+def _prc_img(x: torch.Tensor, t_coord: torch.Tensor | None = None, l_coord: torch.Tensor | None = None):
+    _, h, w = x.shape
+    x_coords = {
+        "t": torch.arange(1, device=x.device) if t_coord is None else t_coord,
+        "h": torch.arange(h, device=x.device),
+        "w": torch.arange(w, device=x.device),
+        "l": torch.arange(1, device=x.device) if l_coord is None else l_coord,
+    }
+    x_ids = torch.cartesian_prod(x_coords["t"], x_coords["h"], x_coords["w"], x_coords["l"])
+    x = rearrange(x, "c h w -> (h w) c")
+    return x, x_ids
+
+
+def _batched_prc_img(
+    x: torch.Tensor, t_coord: torch.Tensor | None = None, l_coord: torch.Tensor | None = None
+):
+    results = []
+    for i in range(len(x)):
+        results.append(
+            _prc_img(
+                x[i],
+                t_coord[i] if t_coord is not None else None,
+                l_coord[i] if l_coord is not None else None,
+            )
+        )
+    x_out, x_ids = zip(*results)
+    return torch.stack(x_out), torch.stack(x_ids)
+
+
+def _compress_time(t_ids: torch.Tensor) -> torch.Tensor:
+    t_ids_max = torch.max(t_ids)
+    t_remap = torch.zeros((t_ids_max + 1,), device=t_ids.device, dtype=t_ids.dtype)
+    t_unique_sorted_ids = torch.unique(t_ids, sorted=True)
+    t_remap[t_unique_sorted_ids] = torch.arange(len(t_unique_sorted_ids), device=t_ids.device, dtype=t_ids.dtype)
+    return t_remap[t_ids]
+
+
+def _scatter_ids(x: torch.Tensor, x_ids: torch.Tensor) -> list[torch.Tensor]:
+    x_list = []
+    for data, pos in zip(x, x_ids):
+        _, ch = data.shape
+        t_ids = pos[:, 0].to(torch.int64)
+        h_ids = pos[:, 1].to(torch.int64)
+        w_ids = pos[:, 2].to(torch.int64)
+        t_ids_cmpr = _compress_time(t_ids)
+        t = torch.max(t_ids_cmpr) + 1
+        h = torch.max(h_ids) + 1
+        w = torch.max(w_ids) + 1
+        flat_ids = t_ids_cmpr * w * h + h_ids * w + w_ids
+        out = torch.zeros((t * h * w, ch), device=data.device, dtype=data.dtype)
+        out.scatter_(0, flat_ids.unsqueeze(1).expand(-1, ch), data)
+        x_list.append(rearrange(out, "(t h w) c -> 1 c t h w", t=t, h=h, w=w))
+    return x_list
+
+
+def _scattercat(x: torch.Tensor, x_ids: torch.Tensor) -> torch.Tensor:
+    return torch.cat(_scatter_ids(x, x_ids), 0).squeeze(2)
+
+
+class SelfFlowPipeline(DiffusionPipeline):
+    """
+    Pipeline for class-conditional Self-Flow image generation on ImageNet 256×256 latents.
+
+    Default sampling uses 250 SDE steps with classifier-free guidance scale 3.5
+    over flow times `(0.0, 0.7)`.
+    """
+
+    model_cpu_offload_seq = "transformer->vae"
+    _optional_components = ["vae"]
+
+    def __init__(
+        self,
+        transformer,
+        scheduler: KarrasDiffusionSchedulers,
+        vae=None,
+        id2label: Optional[Dict[Union[int, str], str]] = None,
+    ):
+        super().__init__()
+        if scheduler is None:
+            scheduler = FlowMatchEulerDiscreteScheduler(
+                num_train_timesteps=1000,
+                shift=1.0,
+                stochastic_sampling=False,
+            )
+        self.register_modules(transformer=transformer, scheduler=scheduler, vae=vae)
+        self.image_processor = VaeImageProcessor()
+        self._id2label = self._normalize_id2label(id2label)
+        self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = bool(self._id2label)
+
+    def _ensure_labels_loaded(self) -> None:
+        if self._labels_loaded_from_model_index:
+            return
+        loaded = self._read_id2label_from_model_index(getattr(self.config, "_name_or_path", None))
+        if loaded:
+            self._id2label = loaded
+            self.labels = self._build_label2id(self._id2label)
+        self._labels_loaded_from_model_index = True
+
+    @staticmethod
+    def _normalize_id2label(id2label: Optional[Dict[Union[int, str], str]]) -> Dict[int, str]:
+        if not id2label:
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _read_id2label_from_model_index(variant_path: Optional[str]) -> Dict[int, str]:
+        if not variant_path:
+            return {}
+        variant_dir = Path(variant_path).resolve()
+        model_index_path = variant_dir / "model_index.json"
+        if not model_index_path.exists():
+            return {}
+        raw = json.loads(model_index_path.read_text(encoding="utf-8"))
+        id2label = raw.get("id2label")
+        if not isinstance(id2label, dict):
+            return {}
+        return {int(key): value for key, value in id2label.items()}
+
+    @staticmethod
+    def _build_label2id(id2label: Dict[int, str]) -> Dict[str, int]:
+        label2id: Dict[str, int] = {}
+        for class_id, value in id2label.items():
+            for synonym in value.split(","):
+                synonym = synonym.strip()
+                if synonym:
+                    label2id[synonym] = int(class_id)
+        return dict(sorted(label2id.items()))
+
+    @property
+    def id2label(self) -> Dict[int, str]:
+        self._ensure_labels_loaded()
+        return self._id2label
+
+    def get_label_ids(self, label: Union[str, List[str]]) -> List[int]:
+        self._ensure_labels_loaded()
+        label2id = self.labels
+        if not label2id:
+            raise ValueError("No English labels loaded. Ensure `id2label` exists in model_index.json.")
+
+        if isinstance(label, str):
+            label = [label]
+
+        missing = [item for item in label if item not in label2id]
+        if missing:
+            preview = ", ".join(list(label2id.keys())[:8])
+            raise ValueError(f"Unknown English label(s): {missing}. Example valid labels: {preview}, ...")
+        return [label2id[item] for item in label]
+
+    def _normalize_class_labels(
+        self,
+        class_labels: Union[int, str, List[Union[int, str]], torch.LongTensor],
+    ) -> torch.LongTensor:
+        if torch.is_tensor(class_labels):
+            return class_labels.to(device=self._execution_device, dtype=torch.long).reshape(-1)
+
+        if isinstance(class_labels, int):
+            class_label_ids = [class_labels]
+        elif isinstance(class_labels, str):
+            class_label_ids = self.get_label_ids(class_labels)
+        elif class_labels and isinstance(class_labels[0], str):
+            class_label_ids = self.get_label_ids(class_labels)
+        else:
+            class_label_ids = list(class_labels)
+
+        return torch.tensor(class_label_ids, device=self._execution_device, dtype=torch.long).reshape(-1)
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path=None, subfolder=None, **kwargs):
+        repo_root = Path(__file__).resolve().parent
+
+        if pretrained_model_name_or_path in (None, "", "."):
+            variant = repo_root
+        else:
+            variant = Path(pretrained_model_name_or_path)
+            if not variant.is_absolute():
+                candidate = (Path.cwd() / variant).resolve()
+                variant = candidate if candidate.exists() else (repo_root / variant).resolve()
+            if subfolder:
+                variant = variant / subfolder
+
+        model_kwargs = dict(kwargs)
+        inserted: List[str] = []
+
+        def _load_component(folder: str, module_name: str, class_name: str):
+            comp_dir = variant / folder
+            module_path = comp_dir / f"{module_name}.py"
+            has_weights = (comp_dir / "config.json").exists() or (comp_dir / "scheduler_config.json").exists()
+            if not module_path.exists() or not has_weights:
+                return None
+
+            comp_path = str(comp_dir)
+            if comp_path not in sys.path:
+                sys.path.insert(0, comp_path)
+                inserted.append(comp_path)
+
+            module = importlib.import_module(module_name)
+            component_cls = getattr(module, class_name)
+            return component_cls.from_pretrained(str(comp_dir), **model_kwargs)
+
+        try:
+            transformer = _load_component("transformer", "transformer_selfflow", "SelfFlowTransformer2DModel")
+            if transformer is None:
+                raise ValueError(f"No loadable transformer found under {variant}")
+
+            scheduler = cls._load_scheduler_from_variant(variant, model_kwargs)
+
+            vae = None
+            vae_dir = variant / "vae"
+            if vae_dir.exists() and (vae_dir / "config.json").exists():
+                vae = AutoencoderKL.from_pretrained(str(vae_dir), **model_kwargs)
+            elif (variant / "vae_pretrained_model_name_or_path.txt").exists():
+                vae_name = (variant / "vae_pretrained_model_name_or_path.txt").read_text(encoding="utf-8").strip()
+                vae = AutoencoderKL.from_pretrained(vae_name, **model_kwargs)
+
+            id2label = cls._read_id2label_from_model_index(str(variant))
+            pipe = cls(transformer=transformer, scheduler=scheduler, vae=vae, id2label=id2label)
+            if hasattr(pipe, "register_to_config"):
+                pipe.register_to_config(_name_or_path=str(variant))
+            return pipe
+        finally:
+            for comp_path in inserted:
+                if comp_path in sys.path:
+                    sys.path.remove(comp_path)
+
+    @classmethod
+    def _load_scheduler_from_variant(cls, variant: Path, model_kwargs: dict):
+        scheduler_dir = variant / "scheduler"
+        config_path = scheduler_dir / "scheduler_config.json"
+        if not config_path.exists():
+            raise ValueError(f"No scheduler config found under {scheduler_dir}")
+
+        scheduler_entry = None
+        model_index_path = variant / "model_index.json"
+        if model_index_path.exists():
+            scheduler_entry = json.loads(model_index_path.read_text(encoding="utf-8")).get("scheduler")
+
+        if scheduler_entry is None:
+            class_name = json.loads(config_path.read_text(encoding="utf-8")).get("_class_name")
+            if not class_name:
+                raise ValueError(f"Missing `_class_name` in {config_path}")
+            scheduler_entry = ["diffusers", class_name]
+
+        if not isinstance(scheduler_entry, list) or len(scheduler_entry) != 2:
+            raise ValueError(f"Invalid scheduler entry in model_index.json: {scheduler_entry}")
+
+        module_name, class_name = scheduler_entry
+        if module_name == "diffusers":
+            import diffusers.schedulers as schedulers_pkg
+
+            scheduler_cls = getattr(schedulers_pkg, class_name)
+            return scheduler_cls.from_pretrained(str(scheduler_dir), **model_kwargs)
+
+        comp_path = str(scheduler_dir)
+        if comp_path not in sys.path:
+            sys.path.insert(0, comp_path)
+        try:
+            module = importlib.import_module(module_name)
+            scheduler_cls = getattr(module, class_name)
+            return scheduler_cls.from_pretrained(str(scheduler_dir), **model_kwargs)
+        finally:
+            if comp_path in sys.path:
+                sys.path.remove(comp_path)
+
+    @staticmethod
+    def prepare_extra_step_kwargs(
+        scheduler: KarrasDiffusionSchedulers,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]],
+    ) -> Dict[str, Any]:
+        kwargs: Dict[str, Any] = {}
+        step_params = set(inspect.signature(scheduler.step).parameters.keys())
+        if "generator" in step_params:
+            kwargs["generator"] = generator
+        return kwargs
+
+    def _patchify_latents(self, latents: torch.Tensor) -> torch.Tensor:
+        patch_size = self.transformer.config.patch_size
+        return rearrange(
+            latents,
+            "b c (h p1) (w p2) -> b (c p1 p2) h w",
+            p1=patch_size,
+            p2=patch_size,
+        )
+
+    def _unpatchify_latents(self, latents: torch.Tensor) -> torch.Tensor:
+        patch_size = self.transformer.config.patch_size
+        channels = self.transformer.config.in_channels
+        return rearrange(
+            latents,
+            "b (c p1 p2) h w -> b c (h p1) (w p2)",
+            p1=patch_size,
+            p2=patch_size,
+            c=channels,
+        )
+
+    def prepare_token_latents(
+        self,
+        batch_size: int,
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        dtype: Optional[torch.dtype] = None,
+        device: Optional[torch.device] = None,
+    ) -> Tuple[torch.Tensor, torch.Tensor]:
+        patch_size = self.transformer.config.patch_size
+        channels = self.transformer.config.in_channels
+        latent_size = self.transformer.config.input_size
+
+        latents = randn_tensor(
+            (batch_size, channels, latent_size, latent_size),
+            generator=generator,
+            device=device,
+            dtype=dtype,
+        )
+        patched = self._patchify_latents(latents)
+        tokens, token_ids = _batched_prc_img(patched)
+        return tokens.to(device=device, dtype=dtype), token_ids.to(device=device)
+
+    @staticmethod
+    def _apply_classifier_free_guidance(
+        model_output: torch.Tensor,
+        guidance_scale: float,
+    ) -> torch.Tensor:
+        if guidance_scale <= 1.0:
+            return model_output
+        model_output_uncond, model_output_cond = model_output.chunk(2)
+        return model_output_uncond + guidance_scale * (model_output_cond - model_output_uncond)
+
+    def decode_latents(self, latents: torch.Tensor, output_type: str = "pil"):
+        if self.vae is None:
+            if output_type == "latent":
+                return latents
+            raise ValueError("Cannot decode latents without a VAE.")
+
+        scaling_factor = getattr(self.vae.config, "scaling_factor", 0.18215)
+        vae_dtype = next(self.vae.parameters()).dtype
+        latents = (latents / scaling_factor).to(dtype=vae_dtype)
+        if output_type == "latent":
+            return latents
+        image = self.vae.decode(latents, return_dict=False)[0]
+        return self.image_processor.postprocess(image, output_type=output_type)
+
+    @torch.inference_mode()
+    def __call__(
+        self,
+        class_labels: Union[int, str, List[Union[int, str]], torch.LongTensor],
+        num_inference_steps: int = 250,
+        guidance_scale: float = 3.5,
+        guidance_interval: Tuple[float, float] = (0.0, 0.7),
+        generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+        output_type: str = "pil",
+        return_dict: bool = True,
+    ) -> Union[ImagePipelineOutput, Tuple]:
+        r"""
+        Generate class-conditional images with Self-Flow.
+
+        Args:
+            class_labels (`int`, `str`, `list[int]`, `list[str]`, or `torch.LongTensor`):
+                ImageNet class indices or human-readable English label strings.
+            num_inference_steps (`int`, defaults to `250`):
+                Number of SDE denoising steps.
+            guidance_scale (`float`, defaults to `3.5`):
+                Classifier-free guidance scale. CFG is active when `guidance_scale > 1.0`
+                and the current flow time lies within `guidance_interval`.
+            guidance_interval (`tuple[float, float]`, defaults to `(0.0, 0.7)`):
+                Flow-time range `(low, high)` where CFG is applied.
+            generator (`torch.Generator`, *optional*):
+                RNG for reproducibility.
+            output_type (`str`, defaults to `"pil"`):
+                `"pil"`, `"np"`, `"pt"`, or `"latent"`.
+            return_dict (`bool`, defaults to `True`):
+                Return [`ImagePipelineOutput`] if True.
+        """
+        device = self._execution_device
+        dtype = next(self.transformer.parameters()).dtype
+
+        class_labels_tensor = self._normalize_class_labels(class_labels)
+        batch_size = class_labels_tensor.shape[0]
+        tokens, token_ids = self.prepare_token_latents(batch_size, generator=generator, dtype=dtype, device=device)
+
+        self.scheduler.set_timesteps(num_inference_steps, device=device)
+        null_labels = torch.full_like(class_labels_tensor, self.transformer.config.num_classes - 1)
+        guidance_low, guidance_high = guidance_interval
+
+        timestep_list = self.scheduler.timesteps.tolist()
+        for timestep_value in self.progress_bar(timestep_list[:-1]):
+            flow_time = float(timestep_value)
+            guidance_active = guidance_scale > 1.0 and guidance_low <= flow_time <= guidance_high
+
+            if guidance_active:
+                model_tokens = torch.cat([tokens, tokens], dim=0)
+                labels = torch.cat([null_labels, class_labels_tensor], dim=0)
+            else:
+                if guidance_scale > 1.0 and tokens.shape[0] > batch_size:
+                    tokens = tokens[batch_size:]
+                model_tokens = tokens
+                labels = class_labels_tensor
+
+            model_t = 1.0 - flow_time
+            timestep_batch = torch.full((model_tokens.shape[0],), model_t, device=device, dtype=dtype)
+            model_output = self.transformer(
+                model_tokens,
+                timestep_batch,
+                labels,
+                return_dict=True,
+            ).sample.to(torch.float32)
+            model_output = -model_output
+
+            if guidance_active:
+                model_output = self._apply_classifier_free_guidance(model_output, guidance_scale)
+                model_output = torch.cat([model_output, model_output], dim=0)
+
+            tokens = self.scheduler.step(
+                model_output,
+                timestep_value,
+                model_tokens,
+                generator=generator,
+            ).prev_sample.to(dtype)
+
+            if guidance_active:
+                tokens = tokens[batch_size:]
+
+        if guidance_scale > 1.0 and tokens.shape[0] > batch_size:
+            tokens = tokens[batch_size:]
+
+        final_time = float(timestep_list[-1])
+        final_model_t = 1.0 - final_time
+        final_timestep = torch.full((tokens.shape[0],), final_model_t, device=device, dtype=dtype)
+        final_output = self.transformer(
+            tokens,
+            final_timestep,
+            class_labels_tensor,
+            return_dict=True,
+        ).sample.to(torch.float32)
+        final_output = -final_output
+        tokens = self.scheduler.step(
+            final_output,
+            final_time,
+            tokens,
+            generator=generator,
+        ).prev_sample.to(dtype)
+
+        spatial = _scattercat(tokens, token_ids)
+        latents = self._unpatchify_latents(spatial)
+        images = self.decode_latents(latents, output_type=output_type)
+
+        self.maybe_free_model_hooks()
+        if not return_dict:
+            return (images,)
+        return ImagePipelineOutput(images=images)
+
+
+SelfFlowPipelineOutput = ImagePipelineOutput

Self-Flow-XL-2-256/scheduler/scheduler_config.json

@@ -0,0 +1,13 @@
+{
+  "_class_name": "SelfFlowFlowMatchScheduler",
+  "_diffusers_version": "0.36.0",
+  "num_train_timesteps": 1000,
+  "path_type": "Linear",
+  "prediction": "velocity",
+  "sampling_method": "Euler",
+  "diffusion_form": "sigma",
+  "diffusion_norm": 1.0,
+  "last_step": "Euler",
+  "last_step_size": 0.04,
+  "reverse": true
+}

Self-Flow-XL-2-256/scheduler/scheduling_flow_match_selfflow.py

@@ -0,0 +1,233 @@
+# Copyright 2026 Black Forest Labs and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+
+from __future__ import annotations
+
+import math
+from dataclasses import dataclass
+from typing import Optional, Tuple, Union
+
+import numpy as np
+import torch
+
+try:
+    from diffusers.configuration_utils import ConfigMixin, register_to_config
+    from diffusers.schedulers.scheduling_utils import SchedulerMixin
+    from diffusers.utils import BaseOutput
+except Exception:  # pragma: no cover
+    class BaseOutput(dict):
+        def __post_init__(self):
+            self.update(self.__dict__)
+
+    class _Config(dict):
+        def __getattr__(self, key):
+            try:
+                return self[key]
+            except KeyError as error:
+                raise AttributeError(key) from error
+
+    class ConfigMixin:
+        config_name = "scheduler_config.json"
+
+    class SchedulerMixin:
+        def set_timesteps(self, *args, **kwargs):
+            raise NotImplementedError
+
+    def register_to_config(init):
+        def wrapper(self, *args, **kwargs):
+            import inspect
+
+            signature = inspect.signature(init)
+            bound = signature.bind(self, *args, **kwargs)
+            bound.apply_defaults()
+            self.config = _Config({key: value for key, value in bound.arguments.items() if key != "self"})
+            init(self, *args, **kwargs)
+
+        return wrapper
+
+
+@dataclass
+class SelfFlowFlowMatchSchedulerOutput(BaseOutput):
+    prev_sample: torch.FloatTensor
+
+
+def _expand_t_like_x(t, x):
+    dims = [1] * (len(x.size()) - 1)
+    return t.view(t.size(0), *dims)
+
+
+class _ICPlan:
+    def compute_alpha_t(self, t):
+        return t, 1
+
+    def compute_sigma_t(self, t):
+        return 1 - t, -1
+
+    def compute_d_alpha_alpha_ratio_t(self, t):
+        return 1 / t
+
+    def compute_diffusion(self, x, t, form="constant", norm=1.0):
+        t = _expand_t_like_x(t, x)
+        choices = {
+            "constant": norm,
+            "SBDM": norm * self._compute_drift(x, t)[1],
+            "sigma": norm * self.compute_sigma_t(t)[0],
+            "linear": norm * (1 - t),
+            "decreasing": 0.25 * (norm * torch.cos(np.pi * t) + 1) ** 2,
+            "increasing-decreasing": norm * torch.sin(np.pi * t) ** 2,
+        }
+        return choices[form]
+
+    def _compute_drift(self, x, t):
+        t = _expand_t_like_x(t, x)
+        alpha_ratio = self.compute_d_alpha_alpha_ratio_t(t)
+        sigma_t, d_sigma_t = self.compute_sigma_t(t)
+        drift = alpha_ratio * x
+        diffusion = alpha_ratio * (sigma_t**2) - sigma_t * d_sigma_t
+        return -drift, diffusion
+
+    def get_score_from_velocity(self, velocity, x, t):
+        t = _expand_t_like_x(t, x)
+        alpha_t, d_alpha_t = self.compute_alpha_t(t)
+        sigma_t, d_sigma_t = self.compute_sigma_t(t)
+        mean = x
+        reverse_alpha_ratio = alpha_t / d_alpha_t
+        var = sigma_t**2 - reverse_alpha_ratio * d_sigma_t * sigma_t
+        score = (reverse_alpha_ratio * velocity - mean) / torch.clamp(var, min=1e-8)
+        # At t=1 the variance vanishes; SDE drift reduces to pure velocity.
+        score = torch.where(var.abs() < 1e-6, torch.zeros_like(score), score)
+        return score
+
+
+class SelfFlowFlowMatchScheduler(SchedulerMixin, ConfigMixin):
+    """
+    Flow-matching SDE scheduler used by Self-Flow ImageNet models.
+    """
+
+    config_name = "scheduler_config.json"
+
+    @register_to_config
+    def __init__(
+        self,
+        num_train_timesteps: int = 1000,
+        path_type: str = "Linear",
+        prediction: str = "velocity",
+        sampling_method: str = "Euler",
+        diffusion_form: str = "sigma",
+        diffusion_norm: float = 1.0,
+        last_step: str = "Mean",
+        last_step_size: float = 0.04,
+        reverse: bool = True,
+    ):
+        self.num_train_timesteps = num_train_timesteps
+        self.path_sampler = _ICPlan()
+        self.timesteps: Optional[torch.Tensor] = None
+        self.sigmas: Optional[torch.Tensor] = None
+        self._step_index = 0
+        self._dt = 1.0
+        self._mean_sample: Optional[torch.Tensor] = None
+
+    def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.device] = None):
+        t0, t1 = 0.0, 1.0
+        timesteps = torch.linspace(t0, t1, num_inference_steps, device=device)
+        self.timesteps = timesteps
+        self.sigmas = 1.0 - timesteps
+        self.num_inference_steps = num_inference_steps
+        self._step_index = 0
+        if len(timesteps) > 1:
+            self._dt = float(timesteps[1] - timesteps[0])
+        else:
+            self._dt = 1.0
+        self._mean_sample = None
+
+    def scale_model_input(self, sample: torch.Tensor, timestep: Union[int, torch.Tensor]) -> torch.Tensor:
+        del timestep
+        return sample
+
+    def _velocity_drift(self, velocity: torch.Tensor) -> torch.Tensor:
+        return velocity
+
+    def _sde_drift(self, sample: torch.Tensor, timestep: torch.Tensor, velocity: torch.Tensor) -> torch.Tensor:
+        diffusion = self.path_sampler.compute_diffusion(
+            sample,
+            timestep,
+            form=self.config.diffusion_form,
+            norm=self.config.diffusion_norm,
+        )
+        score = self.path_sampler.get_score_from_velocity(velocity, sample, timestep)
+        return self._velocity_drift(velocity) + diffusion * score
+
+    def _prepare_timestep(self, timestep: Union[float, torch.Tensor], sample: torch.Tensor) -> torch.Tensor:
+        if not torch.is_tensor(timestep):
+            timestep = torch.tensor([timestep], device=sample.device, dtype=sample.dtype)
+        if timestep.ndim == 0:
+            timestep = timestep.unsqueeze(0)
+        if timestep.numel() == 1:
+            timestep = timestep.expand(sample.shape[0])
+        return timestep.to(device=sample.device, dtype=sample.dtype)
+
+    def step(
+        self,
+        model_output: torch.Tensor,
+        timestep: Union[float, torch.Tensor],
+        sample: torch.Tensor,
+        generator: Optional[torch.Generator] = None,
+        return_dict: bool = True,
+    ) -> Union[SelfFlowFlowMatchSchedulerOutput, Tuple[torch.Tensor]]:
+        if self.timesteps is None:
+            raise ValueError("Call `set_timesteps` before `step`.")
+
+        timestep_tensor = self._prepare_timestep(timestep, sample)
+        velocity = model_output.to(torch.float32)
+        sample = sample.to(torch.float32)
+
+        is_last = self._step_index >= len(self.timesteps) - 1
+        if is_last and self.config.last_step is not None and self.config.last_step_size > 0:
+            drift = self._sde_drift(sample, timestep_tensor, velocity)
+            if self.config.last_step == "Mean":
+                sample = sample + drift * self.config.last_step_size
+            elif self.config.last_step == "Euler":
+                sample = sample + self._velocity_drift(velocity) * self.config.last_step_size
+            self._step_index += 1
+        elif not is_last:
+            if self.config.sampling_method == "Euler":
+                if self._mean_sample is None:
+                    self._mean_sample = sample
+                noise = torch.randn(sample.shape, generator=generator, device=sample.device, dtype=sample.dtype)
+                dw = noise * math.sqrt(self._dt)
+                drift = self._sde_drift(sample, timestep_tensor, velocity)
+                diffusion = self.path_sampler.compute_diffusion(
+                    sample,
+                    timestep_tensor,
+                    form=self.config.diffusion_form,
+                    norm=self.config.diffusion_norm,
+                )
+                self._mean_sample = sample + drift * self._dt
+                sample = self._mean_sample + torch.sqrt(2 * diffusion) * dw
+            elif self.config.sampling_method == "Heun":
+                noise = torch.randn(sample.shape, generator=generator, device=sample.device, dtype=sample.dtype)
+                dw = noise * math.sqrt(self._dt)
+                t_cur = timestep_tensor
+                diffusion = self.path_sampler.compute_diffusion(
+                    sample,
+                    t_cur,
+                    form=self.config.diffusion_form,
+                    norm=self.config.diffusion_norm,
+                )
+                xhat = sample + torch.sqrt(2 * diffusion) * dw
+                k1 = self._sde_drift(xhat, t_cur, velocity)
+                xp = xhat + self._dt * k1
+                k2 = self._sde_drift(xp, t_cur + self._dt, velocity)
+                sample = xhat + 0.5 * self._dt * (k1 + k2)
+                self._mean_sample = xhat
+            else:
+                raise NotImplementedError(f"Sampling method {self.config.sampling_method} is not implemented.")
+            self._step_index += 1
+        else:
+            self._step_index += 1
+
+        prev_sample = sample
+        if not return_dict:
+            return (prev_sample,)
+        return SelfFlowFlowMatchSchedulerOutput(prev_sample=prev_sample)

Self-Flow-XL-2-256/transformer/config.json

@@ -0,0 +1,14 @@
+{
+  "_class_name": "SelfFlowTransformer2DModel",
+  "class_dropout_prob": 0.0,
+  "depth": 28,
+  "hidden_size": 1152,
+  "in_channels": 4,
+  "input_size": 32,
+  "learn_sigma": true,
+  "mlp_ratio": 4.0,
+  "num_classes": 1001,
+  "num_heads": 16,
+  "patch_size": 2,
+  "per_token_timestep": true
+}

Self-Flow-XL-2-256/transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f78e3ba96474e7466f20aa8af40ffaa88248fb61afb19158606357565b530359
+size 2721795664

Self-Flow-XL-2-256/transformer/transformer_selfflow.py

@@ -0,0 +1,412 @@
+# Copyright 2026 Black Forest Labs and The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+
+from __future__ import annotations
+
+import collections.abc
+import math
+from dataclasses import dataclass
+from itertools import repeat
+from typing import Optional, Tuple, Union
+
+import numpy as np
+import torch
+import torch.nn as nn
+from einops import rearrange
+from timm.models.vision_transformer import Attention, Mlp
+
+try:
+    from diffusers.configuration_utils import ConfigMixin, register_to_config
+    from diffusers.models.modeling_utils import ModelMixin
+    from diffusers.utils import BaseOutput
+except Exception:  # pragma: no cover
+    class BaseOutput(dict):
+        def __post_init__(self):
+            self.update(self.__dict__)
+
+    class _Config(dict):
+        def __getattr__(self, key):
+            try:
+                return self[key]
+            except KeyError as error:
+                raise AttributeError(key) from error
+
+    class ConfigMixin:
+        config_name = "config.json"
+
+    class ModelMixin(nn.Module):
+        pass
+
+    def register_to_config(init):
+        def wrapper(self, *args, **kwargs):
+            import inspect
+
+            signature = inspect.signature(init)
+            bound = signature.bind(self, *args, **kwargs)
+            bound.apply_defaults()
+            self.config = _Config({key: value for key, value in bound.arguments.items() if key != "self"})
+            init(self, *args, **kwargs)
+
+        return wrapper
+
+
+def _ntuple(n):
+    def parse(x):
+        if isinstance(x, collections.abc.Iterable) and not isinstance(x, str):
+            return tuple(x)
+        return tuple(repeat(x, n))
+
+    return parse
+
+
+to_2tuple = _ntuple(2)
+
+
+@dataclass
+class SelfFlowTransformer2DModelOutput(BaseOutput):
+    sample: torch.FloatTensor
+
+
+class PatchedPatchEmbed(nn.Module):
+    def __init__(
+        self,
+        img_size: int = 224,
+        patch_size: int = 16,
+        in_chans: int = 3,
+        embed_dim: int = 768,
+        bias: bool = True,
+    ):
+        super().__init__()
+        self.patch_size = to_2tuple(patch_size)
+        img_size = to_2tuple(img_size)
+        self.grid_size = (
+            img_size[0] // self.patch_size[0],
+            img_size[1] // self.patch_size[1],
+        )
+        self.num_patches = self.grid_size[0] * self.grid_size[1]
+        patch_dim = self.patch_size[0] * self.patch_size[1] * in_chans
+        self.proj = nn.Linear(patch_dim, embed_dim, bias=bias)
+
+    def forward(self, x: torch.Tensor) -> torch.Tensor:
+        return self.proj(x)
+
+
+def modulate(x, shift, scale):
+    return x * (1 + scale.unsqueeze(1)) + shift.unsqueeze(1)
+
+
+def modulate_per_token(x, shift, scale):
+    return x * (1 + scale) + shift
+
+
+class TimestepEmbedder(nn.Module):
+    def __init__(self, hidden_size, frequency_embedding_size=256):
+        super().__init__()
+        self.mlp = nn.Sequential(
+            nn.Linear(frequency_embedding_size, hidden_size, bias=True),
+            nn.SiLU(),
+            nn.Linear(hidden_size, hidden_size, bias=True),
+        )
+        self.frequency_embedding_size = frequency_embedding_size
+
+    @staticmethod
+    def timestep_embedding(t, dim, max_period=10000):
+        half = dim // 2
+        freqs = torch.exp(
+            -math.log(max_period)
+            * torch.arange(start=0, end=half, dtype=torch.float32, device=t.device)
+            / half
+        )
+        args = t[:, None].float() * freqs[None]
+        embedding = torch.cat([torch.cos(args), torch.sin(args)], dim=-1)
+        if dim % 2:
+            embedding = torch.cat([embedding, torch.zeros_like(embedding[:, :1])], dim=-1)
+        return embedding
+
+    def forward(self, t):
+        t_freq = self.timestep_embedding(t, self.frequency_embedding_size)
+        t_freq = t_freq.to(dtype=self.mlp[0].weight.dtype)
+        return self.mlp(t_freq)
+
+
+class LabelEmbedder(nn.Module):
+    def __init__(self, num_classes, hidden_size, dropout_prob):
+        super().__init__()
+        use_cfg_embedding = dropout_prob > 0
+        self.embedding_table = nn.Embedding(num_classes + use_cfg_embedding, hidden_size)
+        self.num_classes = num_classes
+        self.dropout_prob = dropout_prob
+
+    def token_drop(self, labels, force_drop_ids=None):
+        if force_drop_ids is None:
+            drop_ids = torch.rand(labels.shape[0], device=labels.device) < self.dropout_prob
+        else:
+            drop_ids = force_drop_ids == 1
+        return torch.where(drop_ids, self.num_classes, labels)
+
+    def forward(self, labels, train, force_drop_ids=None):
+        use_dropout = self.dropout_prob > 0
+        if (train and use_dropout) or (force_drop_ids is not None):
+            labels = self.token_drop(labels, force_drop_ids)
+        return self.embedding_table(labels)
+
+
+class DiTBlock(nn.Module):
+    def __init__(self, hidden_size, num_heads, mlp_ratio=4.0, **block_kwargs):
+        super().__init__()
+        self.norm1 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
+        self.attn = Attention(hidden_size, num_heads=num_heads, qkv_bias=True, **block_kwargs)
+        self.norm2 = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
+        mlp_hidden_dim = int(hidden_size * mlp_ratio)
+        approx_gelu = lambda: nn.GELU(approximate="tanh")
+        self.mlp = Mlp(
+            in_features=hidden_size,
+            hidden_features=mlp_hidden_dim,
+            act_layer=approx_gelu,
+            drop=0,
+        )
+        self.adaLN_modulation = nn.Sequential(
+            nn.SiLU(), nn.Linear(hidden_size, 6 * hidden_size, bias=True)
+        )
+
+    def forward(self, x, c):
+        shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = self.adaLN_modulation(c).chunk(
+            6, dim=1
+        )
+        x = x + gate_msa.unsqueeze(1) * self.attn(modulate(self.norm1(x), shift_msa, scale_msa))
+        x = x + gate_mlp.unsqueeze(1) * self.mlp(modulate(self.norm2(x), shift_mlp, scale_mlp))
+        return x
+
+
+class PerTokenDiTBlock(DiTBlock):
+    def forward(self, x, c):
+        batch_size, seq_len, hidden_dim = c.shape
+        c_flat = c.reshape(-1, hidden_dim)
+        modulation_flat = self.adaLN_modulation(c_flat)
+        modulation = modulation_flat.reshape(batch_size, seq_len, -1)
+        shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = modulation.chunk(6, dim=-1)
+        x = x + gate_msa * self.attn(modulate_per_token(self.norm1(x), shift_msa, scale_msa))
+        x = x + gate_mlp * self.mlp(modulate_per_token(self.norm2(x), shift_mlp, scale_mlp))
+        return x
+
+
+class FinalLayer(nn.Module):
+    def __init__(self, hidden_size, patch_size, out_channels):
+        super().__init__()
+        self.norm_final = nn.LayerNorm(hidden_size, elementwise_affine=False, eps=1e-6)
+        self.linear = nn.Linear(hidden_size, patch_size * patch_size * out_channels, bias=True)
+        self.adaLN_modulation = nn.Sequential(
+            nn.SiLU(), nn.Linear(hidden_size, 2 * hidden_size, bias=True)
+        )
+
+    def forward(self, x, c):
+        shift, scale = self.adaLN_modulation(c).chunk(2, dim=1)
+        x = modulate(self.norm_final(x), shift, scale)
+        return self.linear(x)
+
+
+class PerTokenFinalLayer(FinalLayer):
+    def forward(self, x, c):
+        batch_size, seq_len, hidden_dim = c.shape
+        c_flat = c.reshape(-1, hidden_dim)
+        modulation_flat = self.adaLN_modulation(c_flat)
+        modulation = modulation_flat.reshape(batch_size, seq_len, -1)
+        shift, scale = modulation.chunk(2, dim=-1)
+        x = modulate_per_token(self.norm_final(x), shift, scale)
+        return self.linear(x)
+
+
+class SimpleHead(nn.Module):
+    def __init__(self, in_dim, out_dim):
+        super().__init__()
+        self.linear1 = nn.Linear(in_dim, in_dim + out_dim)
+        self.linear2 = nn.Linear(in_dim + out_dim, out_dim)
+        self.act = nn.SiLU()
+
+    def forward(self, x):
+        x = self.linear1(x)
+        return self.linear2(self.act(x))
+
+
+def get_2d_sincos_pos_embed(embed_dim, grid_size, cls_token=False, extra_tokens=0):
+    grid_h = np.arange(grid_size, dtype=np.float32)
+    grid_w = np.arange(grid_size, dtype=np.float32)
+    grid = np.meshgrid(grid_w, grid_h)
+    grid = np.stack(grid, axis=0)
+    grid = grid.reshape([2, 1, grid_size, grid_size])
+    pos_embed = get_2d_sincos_pos_embed_from_grid(embed_dim, grid)
+    if cls_token and extra_tokens > 0:
+        pos_embed = np.concatenate([np.zeros([extra_tokens, embed_dim]), pos_embed], axis=0)
+    return pos_embed
+
+
+def get_2d_sincos_pos_embed_from_grid(embed_dim, grid):
+    assert embed_dim % 2 == 0
+    emb_h = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[0])
+    emb_w = get_1d_sincos_pos_embed_from_grid(embed_dim // 2, grid[1])
+    return np.concatenate([emb_h, emb_w], axis=1)
+
+
+def get_1d_sincos_pos_embed_from_grid(embed_dim, pos):
+    assert embed_dim % 2 == 0
+    omega = np.arange(embed_dim // 2, dtype=np.float64)
+    omega /= embed_dim / 2.0
+    omega = 1.0 / 10000**omega
+    pos = pos.reshape(-1)
+    out = np.einsum("m,d->md", pos, omega)
+    return np.concatenate([np.sin(out), np.cos(out)], axis=1)
+
+
+class SelfFlowTransformer2DModel(ModelMixin, ConfigMixin):
+    """
+    Self-Flow diffusion transformer with per-token timestep conditioning (SiT-XL/2).
+    """
+
+    config_name = "config.json"
+
+    @register_to_config
+    def __init__(
+        self,
+        input_size: int = 32,
+        patch_size: int = 2,
+        in_channels: int = 4,
+        hidden_size: int = 1152,
+        depth: int = 28,
+        num_heads: int = 16,
+        mlp_ratio: float = 4.0,
+        num_classes: int = 1001,
+        class_dropout_prob: float = 0.0,
+        learn_sigma: bool = True,
+        per_token_timestep: bool = True,
+    ):
+        super().__init__()
+        self.learn_sigma = learn_sigma
+        self.in_channels = in_channels
+        self.out_channels = in_channels * 2 if learn_sigma else in_channels
+        self.patch_size = patch_size
+        self.num_heads = num_heads
+        self.per_token_timestep = per_token_timestep
+
+        self.x_embedder = PatchedPatchEmbed(input_size, patch_size, in_channels, hidden_size, bias=True)
+        self.t_embedder = TimestepEmbedder(hidden_size)
+        self.y_embedder = LabelEmbedder(num_classes, hidden_size, class_dropout_prob)
+
+        num_patches = self.x_embedder.num_patches
+        self.pos_embed = nn.Parameter(torch.zeros(1, num_patches, hidden_size), requires_grad=False)
+
+        block_cls = PerTokenDiTBlock if per_token_timestep else DiTBlock
+        self.blocks = nn.ModuleList(
+            [block_cls(hidden_size, num_heads, mlp_ratio=mlp_ratio) for _ in range(depth)]
+        )
+        final_cls = PerTokenFinalLayer if per_token_timestep else FinalLayer
+        self.final_layer = final_cls(hidden_size, patch_size, self.out_channels)
+        self.projector = SimpleHead(hidden_size, hidden_size)
+
+        self.initialize_weights()
+
+    def initialize_weights(self):
+        def _basic_init(module):
+            if isinstance(module, nn.Linear):
+                torch.nn.init.xavier_uniform_(module.weight)
+                if module.bias is not None:
+                    nn.init.constant_(module.bias, 0)
+
+        self.apply(_basic_init)
+
+        pos_embed = get_2d_sincos_pos_embed(
+            self.pos_embed.shape[-1], int(self.x_embedder.num_patches**0.5)
+        )
+        self.pos_embed.data.copy_(torch.from_numpy(pos_embed).float().unsqueeze(0))
+
+        w = self.x_embedder.proj.weight.data
+        nn.init.xavier_uniform_(w.view([w.shape[0], -1]))
+        nn.init.constant_(self.x_embedder.proj.bias, 0)
+
+        nn.init.normal_(self.y_embedder.embedding_table.weight, std=0.02)
+        nn.init.normal_(self.t_embedder.mlp[0].weight, std=0.02)
+        nn.init.normal_(self.t_embedder.mlp[2].weight, std=0.02)
+
+        for block in self.blocks:
+            nn.init.constant_(block.adaLN_modulation[-1].weight, 0)
+            nn.init.constant_(block.adaLN_modulation[-1].bias, 0)
+
+        nn.init.constant_(self.final_layer.adaLN_modulation[-1].weight, 0)
+        nn.init.constant_(self.final_layer.adaLN_modulation[-1].bias, 0)
+        nn.init.constant_(self.final_layer.linear.weight, 0)
+        nn.init.constant_(self.final_layer.linear.bias, 0)
+
+    def shufflechannel(self, x):
+        p = self.x_embedder.patch_size[0]
+        x = rearrange(x, "b l (p q c) -> b l (c p q)", p=p, q=p, c=self.out_channels)
+        if self.learn_sigma:
+            x, _ = x.chunk(2, dim=2)
+        return x
+
+    def _embed_forward(
+        self,
+        hidden_states: torch.Tensor,
+        timestep: torch.Tensor,
+        class_labels: torch.Tensor,
+    ) -> torch.Tensor:
+        x = self.x_embedder(hidden_states) + self.pos_embed
+        batch_size, seq_len, _ = x.shape
+
+        if self.config.per_token_timestep:
+            if timestep.ndim == 1:
+                t_emb = self.t_embedder(timestep).unsqueeze(1).expand(-1, seq_len, -1)
+            elif timestep.ndim == 2:
+                t_flat = timestep.reshape(-1)
+                t_emb = self.t_embedder(t_flat).reshape(batch_size, seq_len, -1)
+            else:
+                raise ValueError(f"Timesteps must be 1D or 2D, got shape {timestep.shape}")
+            y_emb = self.y_embedder(class_labels, self.training).unsqueeze(1).expand(-1, seq_len, -1)
+            conditioning = t_emb + y_emb
+            for block in self.blocks:
+                x = block(x, conditioning)
+            x = self.final_layer(x, conditioning)
+        else:
+            t_emb = self.t_embedder(timestep)
+            y_emb = self.y_embedder(class_labels, self.training)
+            conditioning = t_emb + y_emb
+            for block in self.blocks:
+                x = block(x, conditioning)
+            x = self.final_layer(x, conditioning)
+        return x
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        timestep: Union[torch.Tensor, float],
+        class_labels: torch.LongTensor,
+        return_dict: bool = True,
+    ) -> Union[SelfFlowTransformer2DModelOutput, torch.Tensor]:
+        """
+        Args:
+            hidden_states: Token tensor `(batch, seq_len, patch_dim)` with `patch_dim = in_channels * patch_size^2`.
+            timestep: Flow time in `[0, 1]` along the SDE path (noise at 0, data at 1).
+                The pipeline passes `1 - flow_time` so that, together with the internal
+                `1 - t` remap and output negation, the effective embedder time matches
+                the official Self-Flow `reverse=True` sampling convention.
+            class_labels: ImageNet class indices (use `num_classes` index for unconditional).
+        """
+        if not torch.is_tensor(timestep):
+            timestep = torch.tensor([timestep], device=hidden_states.device, dtype=hidden_states.dtype)
+        timestep = timestep.to(device=hidden_states.device, dtype=hidden_states.dtype)
+        if timestep.ndim == 1 and timestep.numel() == 1 and hidden_states.shape[0] > 1:
+            timestep = timestep.expand(hidden_states.shape[0])
+
+        # Match legacy inference convention: model expects noise decreasing from 1 -> 0.
+        timestep = 1 - timestep
+
+        class_labels = class_labels.to(device=hidden_states.device, dtype=torch.long).reshape(-1)
+        if class_labels.numel() == 1 and hidden_states.shape[0] > 1:
+            class_labels = class_labels.expand(hidden_states.shape[0])
+
+        sample = self._embed_forward(hidden_states, timestep, class_labels)
+        sample = self.shufflechannel(sample)
+        sample = -sample
+
+        if not return_dict:
+            return sample
+        return SelfFlowTransformer2DModelOutput(sample=sample)

Self-Flow-XL-2-256/vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.38.0",
+  "_name_or_path": "stabilityai/sd-vae-ft-ema",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "shift_factor": null,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": true,
+  "use_quant_conv": true
+}

Self-Flow-XL-2-256/vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:703abdcd7c389316b5128faa9b750a530ea1680b453170b27afebac5e4db30c4
+size 334643268