Upload StableDiffusionXLPipeline

model_index.json

@@ -1,32 +1,36 @@
 {
-  "_class_name": "StableDiffusionInpaintPipeline",
+  "_class_name": "StableDiffusionXLPipeline",
   "_diffusers_version": "0.27.2",
-  "_name_or_path": "/content/Orion/Logs",
+  "_name_or_path": "stabilityai/stable-diffusion-xl-base-1.0",
   "feature_extractor": [
-    "transformers",
-    "CLIPImageProcessor"
-  ],
-  "image_encoder": [
     null,
     null
   ],
-  "requires_safety_checker": false,
-  "safety_checker": [
+  "force_zeros_for_empty_prompt": true,
+  "image_encoder": [
     null,
     null
   ],
   "scheduler": [
     "diffusers",
-    "PNDMScheduler"
+    "EulerDiscreteScheduler"
   ],
   "text_encoder": [
     "transformers",
     "CLIPTextModel"
   ],
+  "text_encoder_2": [
+    "transformers",
+    "CLIPTextModelWithProjection"
+  ],
   "tokenizer": [
     "transformers",
     "CLIPTokenizer"
   ],
+  "tokenizer_2": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
   "unet": [
     "diffusers",
     "UNet2DConditionModel"

scheduler/scheduler_config.json

@@ -1,15 +1,22 @@
 {
-  "_class_name": "PNDMScheduler",
+  "_class_name": "EulerDiscreteScheduler",
   "_diffusers_version": "0.27.2",
   "beta_end": 0.012,
   "beta_schedule": "scaled_linear",
   "beta_start": 0.00085,
   "clip_sample": false,
+  "interpolation_type": "linear",
   "num_train_timesteps": 1000,
   "prediction_type": "epsilon",
+  "rescale_betas_zero_snr": false,
+  "sample_max_value": 1.0,
   "set_alpha_to_one": false,
+  "sigma_max": null,
+  "sigma_min": null,
   "skip_prk_steps": true,
   "steps_offset": 1,
   "timestep_spacing": "leading",
-  "trained_betas": null
+  "timestep_type": "discrete",
+  "trained_betas": null,
+  "use_karras_sigmas": false
 }

text_encoder/config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/content/Orion/Logs/text_encoder",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/text_encoder",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -7,19 +7,19 @@
   "bos_token_id": 0,
   "dropout": 0.0,
   "eos_token_id": 2,
-  "hidden_act": "gelu",
-  "hidden_size": 1024,
+  "hidden_act": "quick_gelu",
+  "hidden_size": 768,
   "initializer_factor": 1.0,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 77,
   "model_type": "clip_text_model",
-  "num_attention_heads": 16,
-  "num_hidden_layers": 23,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
   "pad_token_id": 1,
-  "projection_dim": 512,
+  "projection_dim": 768,
   "torch_dtype": "float16",
-  "transformers_version": "4.41.1",
+  "transformers_version": "4.41.2",
   "vocab_size": 49408
 }

text_encoder/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc1827c465450322616f06dea41596eac7d493f4e95904dcb51f0fc745c4e13f
-size 680820392
+oid sha256:660c6f5b1abae9dc498ac2d21e1347d2abdb0cf6c0c0c8576cd796491d9a6cdd
+size 246144152

text_encoder_2/config.json

@@ -0,0 +1,25 @@
+{
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/text_encoder_2",
+  "architectures": [
+    "CLIPTextModelWithProjection"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "dropout": 0.0,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_size": 1280,
+  "initializer_factor": 1.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 77,
+  "model_type": "clip_text_model",
+  "num_attention_heads": 20,
+  "num_hidden_layers": 32,
+  "pad_token_id": 1,
+  "projection_dim": 1280,
+  "torch_dtype": "float16",
+  "transformers_version": "4.41.2",
+  "vocab_size": 49408
+}

text_encoder_2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec310df2af79c318e24d20511b601a591ca8cd4f1fce1d8dff822a356bcdb1f4
+size 1389382176

tokenizer/special_tokens_map.json

@@ -13,13 +13,7 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": {
-    "content": "!",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
+  "pad_token": "<|endoftext|>",
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer/tokenizer_config.json

@@ -1,14 +1,6 @@
 {
   "add_prefix_space": false,
   "added_tokens_decoder": {
-    "0": {
-      "content": "!",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
     "49406": {
       "content": "<|startoftext|>",
       "lstrip": false,
@@ -32,7 +24,7 @@
   "eos_token": "<|endoftext|>",
   "errors": "replace",
   "model_max_length": 77,
-  "pad_token": "!",
+  "pad_token": "<|endoftext|>",
   "tokenizer_class": "CLIPTokenizer",
   "unk_token": "<|endoftext|>"
 }

tokenizer_2/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<|startoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "!",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_2/tokenizer_config.json

@@ -0,0 +1,38 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "!",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49406": {
+      "content": "<|startoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49407": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": true,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 77,
+  "pad_token": "!",
+  "tokenizer_class": "CLIPTokenizer",
+  "unk_token": "<|endoftext|>"
+}

unet/config.json

@@ -1,22 +1,20 @@
 {
   "_class_name": "UNet2DConditionModel",
   "_diffusers_version": "0.27.2",
-  "_name_or_path": "/content/Orion/Logs/unet",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/unet",
   "act_fn": "silu",
-  "addition_embed_type": null,
+  "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
-  "addition_time_embed_dim": null,
+  "addition_time_embed_dim": 256,
   "attention_head_dim": [
     5,
     10,
-    20,
     20
   ],
   "attention_type": "default",
   "block_out_channels": [
     320,
     640,
-    1280,
     1280
   ],
   "center_input_sample": false,
@@ -24,13 +22,12 @@
   "class_embeddings_concat": false,
   "conv_in_kernel": 3,
   "conv_out_kernel": 3,
-  "cross_attention_dim": 1024,
+  "cross_attention_dim": 2048,
   "cross_attention_norm": null,
   "down_block_types": [
+    "DownBlock2D",
     "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "CrossAttnDownBlock2D",
-    "DownBlock2D"
+    "CrossAttnDownBlock2D"
   ],
   "downsample_padding": 1,
   "dropout": 0.0,
@@ -39,7 +36,7 @@
   "encoder_hid_dim_type": null,
   "flip_sin_to_cos": true,
   "freq_shift": 0,
-  "in_channels": 9,
+  "in_channels": 4,
   "layers_per_block": 2,
   "mid_block_only_cross_attention": null,
   "mid_block_scale_factor": 1,
@@ -50,24 +47,27 @@
   "num_class_embeds": null,
   "only_cross_attention": false,
   "out_channels": 4,
-  "projection_class_embeddings_input_dim": null,
+  "projection_class_embeddings_input_dim": 2816,
   "resnet_out_scale_factor": 1.0,
   "resnet_skip_time_act": false,
   "resnet_time_scale_shift": "default",
   "reverse_transformer_layers_per_block": null,
-  "sample_size": 64,
+  "sample_size": 128,
   "time_cond_proj_dim": null,
   "time_embedding_act_fn": null,
   "time_embedding_dim": null,
   "time_embedding_type": "positional",
   "timestep_post_act": null,
-  "transformer_layers_per_block": 1,
+  "transformer_layers_per_block": [
+    1,
+    2,
+    10
+  ],
   "up_block_types": [
-    "UpBlock2D",
     "CrossAttnUpBlock2D",
     "CrossAttnUpBlock2D",
-    "CrossAttnUpBlock2D"
+    "UpBlock2D"
   ],
-  "upcast_attention": false,
+  "upcast_attention": null,
   "use_linear_projection": true
 }

unet/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ece4bb68f71960f1d47d334620e1460d181c45dfe5ac17a08351cd2844de701
-size 1731933536
+oid sha256:de8a229d07acb65d8edf22925337c687b4089172f2279835edaf6bc0c33e284d
+size 5146944008

vae/config.json

@@ -1,7 +1,7 @@
 {
   "_class_name": "AutoencoderKL",
   "_diffusers_version": "0.27.2",
-  "_name_or_path": "/content/Orion/Logs/vae",
+  "_name_or_path": "/root/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/462165984030d82259a11f4367a4eed129e94a7b/vae",
   "act_fn": "silu",
   "block_out_channels": [
     128,
@@ -23,8 +23,8 @@
   "layers_per_block": 2,
   "norm_num_groups": 32,
   "out_channels": 3,
-  "sample_size": 512,
-  "scaling_factor": 0.18215,
+  "sample_size": 1024,
+  "scaling_factor": 0.13025,
   "up_block_types": [
     "UpDecoderBlock2D",
     "UpDecoderBlock2D",

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e4c08995484ee61270175e9e7a072b66a6e4eeb5f0c266667fe1f45b90daf9a
+oid sha256:bcb60880a46b63dea58e9bc591abe15f8350bde47b405f9c38f4be70c6161e68
 size 167335342