Upload StableCascadeDecoderPipeline

decoder/config.json

@@ -0,0 +1,84 @@
+{
+  "_class_name": "StableCascadeUNet",
+  "_diffusers_version": "0.30.2",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-cascade/snapshots/a89f66d459ae653e3b4d4f992a7c3789d0dc4d16/decoder",
+  "block_out_channels": [
+    320,
+    640,
+    1280,
+    1280
+  ],
+  "block_types_per_layer": [
+    [
+      "SDCascadeResBlock",
+      "SDCascadeTimestepBlock"
+    ],
+    [
+      "SDCascadeResBlock",
+      "SDCascadeTimestepBlock"
+    ],
+    [
+      "SDCascadeResBlock",
+      "SDCascadeTimestepBlock",
+      "SDCascadeAttnBlock"
+    ],
+    [
+      "SDCascadeResBlock",
+      "SDCascadeTimestepBlock",
+      "SDCascadeAttnBlock"
+    ]
+  ],
+  "clip_image_in_channels": null,
+  "clip_seq": 4,
+  "clip_text_in_channels": null,
+  "clip_text_pooled_in_channels": 1280,
+  "conditioning_dim": 1280,
+  "down_blocks_repeat_mappers": [
+    1,
+    1,
+    1,
+    1
+  ],
+  "down_num_layers_per_block": [
+    2,
+    6,
+    28,
+    6
+  ],
+  "dropout": [
+    0,
+    0,
+    0.1,
+    0.1
+  ],
+  "effnet_in_channels": 16,
+  "in_channels": 4,
+  "kernel_size": 3,
+  "num_attention_heads": [
+    0,
+    0,
+    20,
+    20
+  ],
+  "out_channels": 4,
+  "patch_size": 2,
+  "pixel_mapper_in_channels": 3,
+  "self_attn": true,
+  "switch_level": null,
+  "timestep_conditioning_type": [
+    "sca"
+  ],
+  "timestep_ratio_embedding_dim": 64,
+  "up_blocks_repeat_mappers": [
+    3,
+    3,
+    2,
+    2
+  ],
+  "up_num_layers_per_block": [
+    6,
+    28,
+    6,
+    2
+  ]
+}

decoder/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f9575dfa6c2535ad65733d6257d17a7b1e1b54b7eafb251ce9556595f3bc0c9
+size 3126071088

model_index.json

@@ -1,20 +1,12 @@
 {
-  "_class_name": "StableCascadePriorPipeline",
+  "_class_name": "StableCascadeDecoderPipeline",
   "_diffusers_version": "0.30.2",
-  "_name_or_path": "stabilityai/stable-cascade-prior",
-  "feature_extractor": [
-    "transformers",
-    "CLIPImageProcessor"
-  ],
-  "image_encoder": [
-    "transformers",
-    "CLIPVisionModelWithProjection"
-  ],
-  "prior": [
+  "_name_or_path": "stabilityai/stable-cascade",
+  "decoder": [
     "diffusers",
     "StableCascadeUNet"
   ],
-  "resolution_multiple": 42.67,
+  "latent_dim_scale": 10.67,
   "scheduler": [
     "diffusers",
     "DDPMWuerstchenScheduler"
@@ -26,5 +18,9 @@
   "tokenizer": [
     "transformers",
     "CLIPTokenizerFast"
+  ],
+  "vqgan": [
+    "wuerstchen",
+    "PaellaVQModel"
   ]
 }

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-cascade-prior/snapshots/7ca32c21c3b4d4e35bbb94fcfedfb4fa2259bd91/text_encoder",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-cascade/snapshots/a89f66d459ae653e3b4d4f992a7c3789d0dc4d16/text_encoder",
   "architectures": [
     "CLIPTextModelWithProjection"
   ],

vqgan/config.json

@@ -0,0 +1,14 @@
+{
+  "_class_name": "PaellaVQModel",
+  "_diffusers_version": "0.30.2",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-cascade/snapshots/a89f66d459ae653e3b4d4f992a7c3789d0dc4d16/vqgan",
+  "bottleneck_blocks": 12,
+  "embed_dim": 384,
+  "in_channels": 3,
+  "latent_channels": 4,
+  "levels": 2,
+  "num_vq_embeddings": 8192,
+  "out_channels": 3,
+  "scale_factor": 0.3764,
+  "up_down_scale_factor": 2
+}

vqgan/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ac32fab5177329dac907b2480c8c00aeefc712dfd92c2d52263a9c64b426b26
+size 36825828