Upload folder using huggingface_hub

.ipynb_checkpoints/model_index-checkpoint.json

@@ -0,0 +1,32 @@
+{
+  "_class_name": "Kandinsky5T2VPipeline",
+  "_diffusers_version": "0.33.0.dev0",
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "text_encoder": [
+    "transformers",
+    "Qwen2_5_VLForConditionalGeneration"
+  ],
+  "tokenizer": [
+    "transformers",
+    "Qwen2VLProcessor"
+  ],
+  "text_encoder_2": [
+    "transformers",
+    "CLIPTextModel"
+  ],
+  "tokenizer_2": [
+    "transformers",
+    "CLIPTokenizer"
+  ],
+  "transformer": [
+    "diffusers",
+    "Kandinsky5Transformer3DModel"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLHunyuanVideo"
+  ]
+}

transformer/.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,28 @@
+{
+  "_class_name": "Kandinsky5Transformer3DModel",
+  "_diffusers_version": "0.33.0.dev0",
+  "in_visual_dim": 16,
+  "out_visual_dim": 16,
+  "time_dim": 1024,
+  "patch_size": [1, 2, 2],
+  "model_dim": 4096,
+  "ff_dim": 16384,
+  "num_text_blocks": 4,
+  "num_visual_blocks": 60,
+  "axes_dims": [32, 48, 48],
+  "visual_cond": true,
+  "in_text_dim": 3584,
+  "in_text_dim2": 768,
+  "attention_type": "nabla",
+  "attention_chunk": false,
+  "attention_causal": false,
+  "attention_local": false,
+  "attention_glob": false,
+  "attention_window": 3,
+  "attention_method": "topcdf",
+  "attention_P": 0.8,
+  "attention_add_sta": true,
+  "attention_wT": 11,
+  "attention_wW": 3,
+  "attention_wH": 3
+}

transformer/config.json

@@ -0,0 +1,27 @@
+{
+  "_class_name": "Kandinsky5Transformer3DModel",
+  "_diffusers_version": "0.33.0.dev0",
+  "in_visual_dim": 16,
+  "out_visual_dim": 16,
+  "time_dim": 1024,
+  "patch_size": [1, 2, 2],
+  "model_dim": 4096,
+  "ff_dim": 16384,
+  "num_text_blocks": 4,
+  "num_visual_blocks": 60,
+  "axes_dims": [32, 48, 48],
+  "visual_cond": true,
+  "in_text_dim": 3584,
+  "in_text_dim2": 768,
+  "attention_type": "nabla",
+  "attention_causal": false,
+  "attention_local": false,
+  "attention_glob": false,
+  "attention_window": 3,
+  "attention_method": "topcdf",
+  "attention_P": 0.8,
+  "attention_add_sta": true,
+  "attention_wT": 11,
+  "attention_wW": 3,
+  "attention_wH": 3
+}

transformer/config.txt

@@ -0,0 +1,27 @@
+{
+  "_class_name": "Kandinsky5Transformer3DModel",
+  "_diffusers_version": "0.33.0.dev0",
+  "in_visual_dim": 16,
+  "out_visual_dim": 16,
+  "time_dim": 1024,
+  "patch_size": [1, 2, 2],
+  "model_dim": 4096,
+  "ff_dim": 16384,
+  "num_text_blocks": 4,
+  "num_visual_blocks": 60,
+  "axes_dims": [32, 48, 48],
+  "visual_cond": true,
+  "in_text_dim": 3584,
+  "in_text_dim2": 768,
+  "attention_type": "nabla",
+  "attention_causal": false,
+  "attention_local": false,
+  "attention_glob": false,
+  "attention_window": 3,
+  "attention_method": "topcdf",
+  "attention_P": 0.8,
+  "attention_add_sta": true,
+  "attention_wT": 11,
+  "attention_wW": 3,
+  "attention_wH": 3
+}

transformer/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e167e9dfd4e14a2fb76a4ef3d2571df04a0d219f2f6e9121804b0dbcf5ad17c
+size 77245505744