Upload StableCascadePriorPipeline

model_index.json

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

prior/config.json

@@ -1,10 +1,10 @@
 {
   "_class_name": "StableCascadeUNet",
   "_diffusers_version": "0.30.2",
-  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-cascade-prior/snapshots/7ca32c21c3b4d4e35bbb94fcfedfb4fa2259bd91/prior",
+  "_name_or_path": "stabilityai/stable-cascade-prior",
   "block_out_channels": [
-    2048,
-    2048
+    1536,
+    1536
   ],
   "block_types_per_layer": [
     [
@@ -22,14 +22,14 @@
   "clip_seq": 4,
   "clip_text_in_channels": 1280,
   "clip_text_pooled_in_channels": 1280,
-  "conditioning_dim": 2048,
+  "conditioning_dim": 1536,
   "down_blocks_repeat_mappers": [
     1,
     1
   ],
   "down_num_layers_per_block": [
-    8,
-    24
+    4,
+    12
   ],
   "dropout": [
     0.1,
@@ -39,8 +39,8 @@
   "in_channels": 16,
   "kernel_size": 3,
   "num_attention_heads": [
-    32,
-    32
+    24,
+    24
   ],
   "out_channels": 16,
   "patch_size": 1,
@@ -59,7 +59,7 @@
     1
   ],
   "up_num_layers_per_block": [
-    24,
-    8
+    12,
+    4
   ]
 }

prior/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44a4cd9540f327f2fb4ac09179e4e87912a01cdb1b3b86c79f0f853976fb4c98
-size 7178377816
+oid sha256:61a01756dcaecda654074624fd5f993dbe22c7f5cb0d08887416e6c594179a6a
+size 4123225040

text_encoder/config.json

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

text_encoder/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30e7a50f59884f068ac6bdc41b0ade553f23c0cbe2eaf3bac76369d2c39bfa37
+oid sha256:fa5b2e6f4c2efc2d82e4b8312faec1a5540eabfc6415126c9a05c8436a530ef4
 size 2778702264