Upload 9 files

.gitattributes

@@ -0,0 +1 @@
+illustriousXL_smoothftSPO.safetensors filter=lfs diff=lfs merge=lfs -text

illustriousXL_smoothftSPO.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac1faa25f64ddd26463653bb303bad0b252262ced084e3040953481cccc18593
+size 6938041160

model_index.json

@@ -0,0 +1,12 @@
+{
+    "_class_name": "StableDiffusionXLPipeline",
+    "architectures": ["StableDiffusionXLPipeline"],
+    "_diffusers_version": "0.30.3",
+    "torch_dtype": "float16",
+    "cross_attention_dim": 2048,
+    "num_train_timesteps": 1000,
+    "transformer_layers_per_block": 10,
+    "projection_class_embeddings_input_dim": 2816,
+    "force_zeros_for_empty_prompt": true,
+    "license": "creativeml-openrail-m"
+}

scheduler/scheduler_config.json

@@ -0,0 +1,12 @@
+{
+    "_class_name": "EulerDiscreteScheduler",
+    "beta_end": 0.012,
+    "beta_schedule": "scaled_linear",
+    "beta_start": 0.00085,
+    "num_train_timesteps": 1000,
+    "steps_offset": 1,
+    "clip_sample": false,
+    "set_alpha_to_one": false,
+    "prediction_type": "epsilon",
+    "trained_betas": null
+}

text_encoder/config.json

@@ -0,0 +1,16 @@
+{
+    "architectures": ["CLIPTextModel"],
+    "hidden_size": 1280,
+    "intermediate_size": 5120,
+    "num_hidden_layers": 24,
+    "num_attention_heads": 16,
+    "max_position_embeddings": 77,
+    "hidden_act": "gelu",
+    "layer_norm_eps": 1e-12,
+    "dropout": 0.1,
+    "vocab_size": 49408,
+    "projection_dim": 1280,
+    "pad_token_id": 1,
+    "bos_token_id": 0,
+    "eos_token_id": 2
+}

text_encoder_2/config.json

@@ -0,0 +1,16 @@
+{
+    "architectures": ["CLIPTextModel"],
+    "hidden_size": 1280,
+    "intermediate_size": 5120,
+    "num_hidden_layers": 24,
+    "num_attention_heads": 16,
+    "max_position_embeddings": 77,
+    "hidden_act": "gelu",
+    "layer_norm_eps": 1e-12,
+    "dropout": 0.1,
+    "vocab_size": 49408,
+    "projection_dim": 1280,
+    "pad_token_id": 1,
+    "bos_token_id": 0,
+    "eos_token_id": 2
+}

tokenizer/tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+    "model_type": "clip",
+    "tokenizer_type": "CLIPTokenizer",
+    "vocab_size": 49408,
+    "max_position_embeddings": 77,
+    "pad_token_id": 1,
+    "bos_token_id": 0,
+    "eos_token_id": 2,
+    "special_tokens": {
+        "pad_token": "<|pad|>",
+        "bos_token": "<|startoftext|>",
+        "eos_token": "<|endoftext|>"
+    }
+}

tokenizer_2/tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+    "model_type": "clip",
+    "tokenizer_type": "CLIPTokenizer",
+    "vocab_size": 49408,
+    "max_position_embeddings": 77,
+    "pad_token_id": 1,
+    "bos_token_id": 0,
+    "eos_token_id": 2,
+    "special_tokens": {
+        "pad_token": "<|pad|>",
+        "bos_token": "<|startoftext|>",
+        "eos_token": "<|endoftext|>"
+    }
+}

unet/config.json

@@ -0,0 +1,26 @@
+{
+    "sample_size": 128,
+    "in_channels": 4,
+    "out_channels": 4,
+    "center_input_sample": false,
+    "flip_sin_to_cos": true,
+    "freq_shift": 0,
+    "down_block_types": [
+        "DownBlock2D",
+        "CrossAttnDownBlock2D",
+        "CrossAttnDownBlock2D"
+    ],
+    "mid_block_type": "UNetMidBlock2DCrossAttn",
+    "up_block_types": [
+        "CrossAttnUpBlock2D",
+        "CrossAttnUpBlock2D",
+        "UpBlock2D"
+    ],
+    "block_out_channels": [320, 640, 1280],
+    "layers_per_block": 2,
+    "cross_attention_dim": 2048,
+    "transformer_layers_per_block": 10,
+    "attention_head_dim": 8,
+    "num_attention_heads": 16,
+    "use_linear_projection": true
+}

vae/config.json

@@ -0,0 +1,19 @@
+{
+    "sample_size": 128,
+    "in_channels": 3,
+    "out_channels": 3,
+    "down_block_types": [
+        "DownEncoderBlock2D",
+        "DownEncoderBlock2D",
+        "DownEncoderBlock2D"
+    ],
+    "up_block_types": [
+        "UpDecoderBlock2D",
+        "UpDecoderBlock2D",
+        "UpDecoderBlock2D"
+    ],
+    "block_out_channels": [128, 256, 512],
+    "layers_per_block": 2,
+    "latent_channels": 4,
+    "scaling_factor": 0.18215
+}