switching models

scheduler/scheduler_config.json

@@ -1,6 +1,6 @@
 {
-  "_class_name": "DDPMScheduler",
-  "_diffusers_version": "0.21.0.dev0",
+  "_class_name": "EulerDiscreteScheduler",
+  "_diffusers_version": "0.19.3",
   "beta_end": 0.012,
   "beta_schedule": "scaled_linear",
   "beta_start": 0.00085,
@@ -14,6 +14,5 @@
   "steps_offset": 1,
   "timestep_spacing": "leading",
   "trained_betas": null,
-  "use_karras_sigmas": false,
-  "rescale_betas_zero_snr": true
+  "use_karras_sigmas": false
 }

text_encoder/config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "/home/suraj_huggingface_co/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/bf714989e22c57ddc1c453bf74dab4521acb81d8/text_encoder",
   "architectures": [
     "CLIPTextModel"
   ],
@@ -19,7 +18,7 @@
   "num_hidden_layers": 12,
   "pad_token_id": 1,
   "projection_dim": 768,
-  "torch_dtype": "float16",
+  "torch_dtype": "float32",
   "transformers_version": "4.29.2",
   "vocab_size": 49408
 }

unet/diffusion_pytorch_model.bin → text_encoder/model.fp16.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:046b775cb9bbc67635fc3b148bb03bfe00496ce2f9ce8488a82fdb388669a521
-size 11965769774
+oid sha256:261b2e72bd3b76723eff6f9857c23cd8f8de6a12129d5351b56271b9202ac0aa
+size 246144864

text_encoder/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79f531155c765c22c89e23328793a2e91a1178070af961c57e2eae5f0509b65b
-size 492265879
+oid sha256:c4cc307234f0bdbf4f078ca9d62efc6d6f58995d25f1be1626db13bf1d5ae0d6
+size 492265880

text_encoder_2/config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "/home/suraj_huggingface_co/.cache/huggingface/hub/models--stabilityai--stable-diffusion-xl-base-1.0/snapshots/bf714989e22c57ddc1c453bf74dab4521acb81d8/text_encoder_2",
   "architectures": [
     "CLIPTextModelWithProjection"
   ],
@@ -19,7 +18,7 @@
   "num_hidden_layers": 32,
   "pad_token_id": 1,
   "projection_dim": 1280,
-  "torch_dtype": "float16",
+  "torch_dtype": "float32",
   "transformers_version": "4.29.2",
   "vocab_size": 49408
 }

text_encoder_2/model.fp16.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de815ea750a4ceeeff4452c6023f894fa66db5db6e7053e589be3104922dd8e2
+size 1389382880

text_encoder_2/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:283bb90f987a133dec11947571aca17692ed32f3fff708441ac8eedcfa4a040e
+oid sha256:12837bb720fa6758ea501af6fc19e8ef95642f409e82833cb9bee97bfc58ece0
 size 2778702976

unet/config.json

@@ -1,7 +1,6 @@
 {
   "_class_name": "UNet2DConditionModel",
-  "_diffusers_version": "0.25.0",
-  "_name_or_path": "valhalla/sdxl-inpaint-ema",
+  "_diffusers_version": "0.19.3",
   "act_fn": "silu",
   "addition_embed_type": "text_time",
   "addition_embed_type_num_heads": 64,
@@ -11,7 +10,6 @@
     10,
     20
   ],
-  "attention_type": "default",
   "block_out_channels": [
     320,
     640,
@@ -24,7 +22,6 @@
   "conv_out_kernel": 3,
   "cross_attention_dim": 2048,
   "cross_attention_norm": null,
-  "decay": 0.9999,
   "down_block_types": [
     "DownBlock2D",
     "CrossAttnDownBlock2D",
@@ -32,25 +29,21 @@
   ],
   "downsample_padding": 1,
   "dual_cross_attention": false,
-  "encoder_hid_dim": 1280,
-  "encoder_hid_dim_type": "ip_image_proj",
+  "encoder_hid_dim": null,
+  "encoder_hid_dim_type": null,
   "flip_sin_to_cos": true,
   "freq_shift": 0,
-  "in_channels": 13,
-  "inv_gamma": 1.0,
+  "in_channels": 4,
   "layers_per_block": 2,
   "mid_block_only_cross_attention": null,
   "mid_block_scale_factor": 1,
   "mid_block_type": "UNetMidBlock2DCrossAttn",
-  "min_decay": 0.0,
   "norm_eps": 1e-05,
   "norm_num_groups": 32,
   "num_attention_heads": null,
   "num_class_embeds": null,
   "only_cross_attention": false,
-  "optimization_step": 37000,
   "out_channels": 4,
-  "power": 0.6666666666666666,
   "projection_class_embeddings_input_dim": 2816,
   "resnet_out_scale_factor": 1.0,
   "resnet_skip_time_act": false,
@@ -72,7 +65,5 @@
     "UpBlock2D"
   ],
   "upcast_attention": null,
-  "update_after_step": 0,
-  "use_ema_warmup": false,
   "use_linear_projection": true
 }

unet/diffusion_pytorch_model.fp16.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc2027ec2344f8554ba3a691130f41707b23bb4822db44463b5434343c856003
+size 5135149760

unet/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8924e82b9cab81e8eadeddc78faee044e99e89ff46a3106e29f1d0c840920bc0
+size 10270077736

vae/config.json

@@ -1,7 +1,6 @@
 {
   "_class_name": "AutoencoderKL",
-  "_diffusers_version": "0.21.0.dev0",
-  "_name_or_path": "madebyollin/sdxl-vae-fp16-fix",
+  "_diffusers_version": "0.19.3",
   "act_fn": "silu",
   "block_out_channels": [
     128,
@@ -15,13 +14,13 @@
     "DownEncoderBlock2D",
     "DownEncoderBlock2D"
   ],
-  "force_upcast": false,
+  "force_upcast": true,
   "in_channels": 3,
   "latent_channels": 4,
   "layers_per_block": 2,
   "norm_num_groups": 32,
   "out_channels": 3,
-  "sample_size": 512,
+  "sample_size": 1024,
   "scaling_factor": 0.13025,
   "up_block_types": [
     "UpDecoderBlock2D",

vae/diffusion_pytorch_model.fp16.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6353737672c94b96174cb590f711eac6edf2fcce5b6e91aa9d73c5adc589ee48
+size 167335342

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:98a14dc6fe8d71c83576f135a87c61a16561c9c080abba418d2cc976ee034f88
+oid sha256:78f6189c8492013e3cac81637a1f657f790a237387f8a9dfd6bfa5fee28eb646
 size 334643268