Upload files with `sana-upload`.

Upload model_index.json
Upload text_encoder/config.json
Upload tokenizer/tokenizer.json
Upload transformer/diffusion_pytorch_model.safetensors.index.json
Upload transformer/config.json
Upload transformer/diffusion_pytorch_model-00001-of-00002.safetensors
Upload transformer/diffusion_pytorch_model-00002-of-00002.safetensors
Upload vae/diffusion_pytorch_model.safetensors
Upload vae/config.json

model_index.json

@@ -19,6 +19,6 @@
   ],
   "vae": [
     "diffusers",
-    "DCAE"
+    "AutoencoderDC"
   ]
 }

text_encoder/config.json

@@ -30,7 +30,7 @@
   "rope_theta": 10000.0,
   "sliding_window": 4096,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.45.2",
+  "transformers_version": "4.46.0",
   "use_cache": true,
   "vocab_size": 256000
 }

tokenizer/tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92a4cfae35c3ce9489bbf9e7a7863334db333e7b3008bf74265617178b93abdd
-size 34363138
+oid sha256:5f7eee611703c5ce5d1eee32d9cdcfe465647b8aff0c1dfb3bed7ad7dbb05060
+size 34362873

transformer/config.json

@@ -1,45 +1,20 @@
 {
   "_class_name": "SanaTransformer2DModel",
   "_diffusers_version": "0.32.0.dev0",
-  "activation_fn": [
-    "silu",
-    "silu",
-    null
-  ],
   "attention_bias": false,
   "attention_head_dim": 32,
-  "attention_type": "default",
   "caption_channels": 2304,
-  "caption_norm_scale_factor": 0.1,
   "cross_attention_dim": 2240,
   "cross_attention_head_dim": 112,
   "dropout": 0.0,
-  "expand_ratio": 2.5,
-  "ff_bias": [
-    true,
-    true,
-    false
-  ],
-  "ff_norm": [
-    null,
-    null,
-    null
-  ],
   "in_channels": 32,
-  "interpolation_scale": null,
+  "mlp_ratio": 2.5,
   "norm_elementwise_affine": false,
   "norm_eps": 1e-06,
-  "norm_num_groups": 32,
-  "norm_type": "ada_norm_single",
   "num_attention_heads": 70,
   "num_cross_attention_heads": 20,
-  "num_embeds_ada_norm": 1000,
   "num_layers": 20,
   "out_channels": 32,
   "patch_size": 1,
-  "sample_size": 32,
-  "upcast_attention": false,
-  "use_additional_conditions": false,
-  "use_caption_norm": true,
-  "use_pe": false
+  "sample_size": 32
 }

transformer/diffusion_pytorch_model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed9a7b95dc53b9ee4adf58513bd7ae312d7c1610128d0578943dfa1e1d708f14
+size 4999769624

transformer/diffusion_pytorch_model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3369a57d76d00d6cfb0d1591640c3fed4c1b60fb1d7ed611fd0cfc79d76874e4
+size 1418127024

transformer/diffusion_pytorch_model.safetensors.index.json

@@ -0,0 +1,403 @@
+{
+  "metadata": {
+    "total_size": 6417851008
+  },
+  "weight_map": {
+    "caption_norm.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "caption_projection.linear_1.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "caption_projection.linear_1.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "caption_projection.linear_2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "caption_projection.linear_2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "patch_embed.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "patch_embed.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "proj_out.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "proj_out.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.emb.timestep_embedder.linear_1.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.emb.timestep_embedder.linear_1.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.emb.timestep_embedder.linear_2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.emb.timestep_embedder.linear_2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.linear.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "time_embed.linear.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.0.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.1.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.10.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.11.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.12.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.13.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.14.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.15.ff.conv_depth.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.ff.conv_depth.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.ff.conv_inverted.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.ff.conv_inverted.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.ff.conv_point.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.15.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.16.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.attn2.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.ff.conv_depth.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.ff.conv_depth.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.ff.conv_inverted.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.ff.conv_inverted.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.ff.conv_point.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.16.scale_shift_table": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.attn2.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.ff.conv_depth.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.ff.conv_depth.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.ff.conv_inverted.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.ff.conv_inverted.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.ff.conv_point.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.17.scale_shift_table": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.attn2.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.ff.conv_depth.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.ff.conv_depth.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.ff.conv_inverted.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.ff.conv_inverted.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.ff.conv_point.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.18.scale_shift_table": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.attn2.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.ff.conv_depth.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.ff.conv_depth.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.ff.conv_inverted.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.ff.conv_inverted.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.ff.conv_point.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.19.scale_shift_table": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "transformer_blocks.2.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.2.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.3.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.4.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.5.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.6.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.7.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.8.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.attn2.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.ff.conv_depth.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.ff.conv_depth.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.ff.conv_inverted.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.ff.conv_inverted.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.ff.conv_point.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "transformer_blocks.9.scale_shift_table": "diffusion_pytorch_model-00001-of-00002.safetensors"
+  }
+}

vae/config.json

@@ -1,17 +1,26 @@
 {
-  "_class_name": "DCAE",
+  "_class_name": "AutoencoderDC",
   "_diffusers_version": "0.32.0.dev0",
-  "_name_or_path": "Efficient-Large-Model/dc_ae_f32c32_sana_1.0_diffusers",
-  "decoder_act": "silu",
-  "decoder_block_type": [
+  "_name_or_path": "mit-han-lab/dc-ae-f32c32-sana-1.0-diffusers",
+  "attention_head_dim": 32,
+  "decoder_act_fns": "silu",
+  "decoder_block_out_channels": [
+    128,
+    256,
+    512,
+    512,
+    1024,
+    1024
+  ],
+  "decoder_block_types": [
     "ResBlock",
     "ResBlock",
     "ResBlock",
-    "EViTS5GLU",
-    "EViTS5GLU",
-    "EViTS5GLU"
+    "EfficientViTBlock",
+    "EfficientViTBlock",
+    "EfficientViTBlock"
   ],
-  "decoder_depth_list": [
+  "decoder_layers_per_block": [
     3,
     3,
     3,
@@ -19,8 +28,23 @@
     3,
     3
   ],
-  "decoder_norm": "rms2d",
-  "decoder_width_list": [
+  "decoder_norm_types": "rms_norm",
+  "decoder_qkv_multiscales": [
+    [],
+    [],
+    [],
+    [
+      5
+    ],
+    [
+      5
+    ],
+    [
+      5
+    ]
+  ],
+  "downsample_block_type": "Conv",
+  "encoder_block_out_channels": [
     128,
     256,
     512,
@@ -28,17 +52,15 @@
     1024,
     1024
   ],
-  "downsample_block_type": "Conv",
-  "encoder_act": "silu",
-  "encoder_block_type": [
+  "encoder_block_types": [
     "ResBlock",
     "ResBlock",
     "ResBlock",
-    "EViTS5GLU",
-    "EViTS5GLU",
-    "EViTS5GLU"
+    "EfficientViTBlock",
+    "EfficientViTBlock",
+    "EfficientViTBlock"
   ],
-  "encoder_depth_list": [
+  "encoder_layers_per_block": [
     2,
     2,
     2,
@@ -46,17 +68,22 @@
     3,
     3
   ],
-  "encoder_norm": "rms2d",
-  "encoder_width_list": [
-    128,
-    256,
-    512,
-    512,
-    1024,
-    1024
+  "encoder_qkv_multiscales": [
+    [],
+    [],
+    [],
+    [
+      5
+    ],
+    [
+      5
+    ],
+    [
+      5
+    ]
   ],
   "in_channels": 3,
   "latent_channels": 32,
   "scaling_factor": 0.41407,
-  "upsample_block_type": "InterpolateConv"
+  "upsample_block_type": "interpolate"
 }

vae/diffusion_pytorch_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25a1d9ac3b3422160ce8a4b5454ed917f103bb18e30fc1b307dec66375167bb8
-size 1249041852
+oid sha256:15a4b09e56d95b768a0ec9da50b702e21d920333fc9b3480d66bb5c7fad9d87f
+size 1249044836