Upload folder using huggingface_hub

lam/config.json

@@ -0,0 +1,61 @@
+{
+  "architectures": [
+    "LAMModel"
+  ],
+  "encoder_height": 224,
+  "encoder_width": 304,
+  "fsq_levels": [
+    7,
+    5,
+    5,
+    5,
+    5
+  ],
+  "initializer_range": 0.02,
+  "is_diffusion": true,
+  "latent_channels": 5,
+  "max_tokens": 80,
+  "min_tokens": 1,
+  "model_type": "lam",
+  "null_latent": 0,
+  "num_tokens_in": 333,
+  "pool_pick_tokens": 80,
+  "torch_dtype": "float32",
+  "transformer_config": {
+    "_class_name": "SD3Transformer2DModel",
+    "_diffusers_version": "0.31.0.dev0",
+    "_name_or_path": "stabilityai/stable-diffusion-3.5-medium",
+    "attention_head_dim": 64,
+    "caption_projection_dim": 1536,
+    "dual_attention_layers": [
+      0,
+      1,
+      2,
+      3,
+      4,
+      5,
+      6,
+      7,
+      8,
+      9,
+      10,
+      11,
+      12
+    ],
+    "in_channels": 32,
+    "joint_attention_dim": 5,
+    "num_attention_heads": 24,
+    "num_layers": 24,
+    "out_channels": 16,
+    "patch_size": 2,
+    "pooled_projection_dim": 400,
+    "pos_embed_max_size": 384,
+    "qk_norm": "rms_norm",
+    "sample_size": 128
+  },
+  "transformers_version": "4.50.0",
+  "use_empty_prompt": true,
+  "use_tail_drop": true,
+  "videomae_from_pretrained": "OpenGVLab/VideoMAEv2-Large",
+  "vocab_size": 4375
+}

lam/model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a368298faeb43922ec189f00f45d674870195f45d6b9974446e7664e3a2a6f12
+size 4989515188

lam/model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:726af20335f395ba445159b3244985493792d307046edc4dd31be79b72d63a23
+size 4995121448

lam/model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a783221322273380a9ce0b454de0da9454f96ff68f3c1b373c31a54136a40a4
+size 1076820760

model_index.json

@@ -0,0 +1,20 @@
+{
+  "_class_name": "LAMPipeline",
+  "_diffusers_version": "0.32.0",
+  "lam": [
+    "vll.models.lam.modeling_lam",
+    "LAMModel"
+  ],
+  "processor": [
+    "vll.models.lam.processing_lam",
+    "LAMProcessor"
+  ],
+  "scheduler": [
+    "diffusers",
+    "FlowMatchEulerDiscreteScheduler"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
+}

processor/processor_config.json

@@ -0,0 +1,8 @@
+{
+  "encoder_height": 224,
+  "encoder_width": 304,
+  "height": 288,
+  "processor_class": "LAMProcessor",
+  "vae_scale_factor": 8,
+  "width": 384
+}

scheduler/scheduler_config.json

@@ -0,0 +1,16 @@
+{
+  "_class_name": "FlowMatchEulerDiscreteScheduler",
+  "_diffusers_version": "0.32.0",
+  "base_image_seq_len": 256,
+  "base_shift": 0.5,
+  "invert_sigmas": false,
+  "max_image_seq_len": 4096,
+  "max_shift": 1.15,
+  "num_train_timesteps": 1000,
+  "shift": 3.0,
+  "shift_terminal": null,
+  "use_beta_sigmas": false,
+  "use_dynamic_shifting": false,
+  "use_exponential_sigmas": false,
+  "use_karras_sigmas": false
+}

vae/config.json

@@ -0,0 +1,38 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.32.0",
+  "_name_or_path": "stabilityai/stable-diffusion-3.5-medium",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "force_upcast": true,
+  "in_channels": 3,
+  "latent_channels": 16,
+  "latents_mean": null,
+  "latents_std": null,
+  "layers_per_block": 2,
+  "mid_block_add_attention": true,
+  "norm_num_groups": 32,
+  "out_channels": 3,
+  "sample_size": 1024,
+  "scaling_factor": 1.5305,
+  "shift_factor": 0.0609,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ],
+  "use_post_quant_conv": false,
+  "use_quant_conv": false
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:965f71f229a5981c0a025db771d8868998f5e4e8412dc5585c687f638fa59679
+size 335306212