Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text

audio_vae/config.json

@@ -0,0 +1,25 @@
+{
+  "_class_name": "AutoencoderKLLTX2Audio",
+  "_diffusers_version": "0.41.0.dev0",
+  "attn_resolutions": null,
+  "base_channels": 128,
+  "causality_axis": "height",
+  "ch_mult": [
+    1,
+    2,
+    4
+  ],
+  "double_z": true,
+  "dropout": 0.0,
+  "in_channels": 2,
+  "is_causal": true,
+  "latent_channels": 8,
+  "mel_bins": 64,
+  "mel_hop_length": 160,
+  "mid_block_add_attention": false,
+  "norm_type": "pixel",
+  "num_res_blocks": 2,
+  "output_channels": 2,
+  "resolution": 256,
+  "sample_rate": 16000
+}

audio_vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45dcb1c3cffc8c8b7710a04dd19500e481daff31f5ea8e8529f2cf697c439d12
+size 106507972

connectors/config.json

@@ -0,0 +1,25 @@
+{
+  "_class_name": "LTX2TextConnectors",
+  "_diffusers_version": "0.41.0.dev0",
+  "audio_connector_attention_head_dim": 64,
+  "audio_connector_num_attention_heads": 32,
+  "audio_connector_num_layers": 8,
+  "audio_connector_num_learnable_registers": 128,
+  "audio_gated_attn": true,
+  "audio_hidden_dim": 2048,
+  "caption_channels": 3840,
+  "causal_temporal_positioning": false,
+  "connector_rope_base_seq_len": 4096,
+  "per_modality_projections": true,
+  "proj_bias": true,
+  "rope_double_precision": true,
+  "rope_theta": 10000.0,
+  "rope_type": "split",
+  "text_proj_in_factor": 49,
+  "video_connector_attention_head_dim": 128,
+  "video_connector_num_attention_heads": 32,
+  "video_connector_num_layers": 8,
+  "video_connector_num_learnable_registers": 128,
+  "video_gated_attn": true,
+  "video_hidden_dim": 4096
+}

connectors/diffusion_pytorch_model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b767a411e336d692306d7d34fa85627db1d17809a4690f45dd96a4b3807d248c
+size 4999998520

connectors/diffusion_pytorch_model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffe33521b2bbd12604abe2d63b5cf02cf55f2fdf10226ff5aae00e437678ebc8
+size 1344490576

connectors/diffusion_pytorch_model.safetensors.index.json

@@ -0,0 +1,269 @@
+{
+  "metadata": {
+    "total_size": 6344455168
+  },
+  "weight_map": {
+    "audio_connector.learnable_registers": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.0.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.1.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.2.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.3.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.4.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.5.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.6.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_connector.transformer_blocks.7.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "audio_text_proj_in.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "audio_text_proj_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.learnable_registers": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.0.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.1.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.2.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.3.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.4.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.ff.net.2.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.5.ff.net.2.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.norm_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.norm_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_k.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_k.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_out.0.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_out.0.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_q.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_q.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_v.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.attn1.to_v.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.ff.net.0.proj.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.ff.net.0.proj.weight": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.6.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.norm_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.norm_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_gate_logits.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_gate_logits.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_k.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_k.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_out.0.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_out.0.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_q.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_q.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_v.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.attn1.to_v.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.ff.net.0.proj.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.ff.net.0.proj.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.ff.net.2.bias": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_connector.transformer_blocks.7.ff.net.2.weight": "diffusion_pytorch_model-00002-of-00002.safetensors",
+    "video_text_proj_in.bias": "diffusion_pytorch_model-00001-of-00002.safetensors",
+    "video_text_proj_in.weight": "diffusion_pytorch_model-00001-of-00002.safetensors"
+  }
+}

modular_model_index.json

@@ -0,0 +1,103 @@
+{
+  "_blocks_class_name": "EchoWMFlashBlocks",
+  "_class_name": "EchoWMFlashModularPipeline",
+  "_diffusers_version": "0.41.0.dev0",
+  "audio_vae": [
+    "diffusers",
+    "AutoencoderKLLTX2Audio",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "audio_vae",
+      "type_hint": [
+        "diffusers",
+        "AutoencoderKLLTX2Audio"
+      ],
+      "variant": null
+    }
+  ],
+  "connectors": [
+    "ltx2",
+    "LTX2TextConnectors",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "connectors",
+      "type_hint": [
+        "ltx2",
+        "LTX2TextConnectors"
+      ],
+      "variant": null
+    }
+  ],
+  "text_encoder": [
+    "transformers",
+    "Gemma3ForConditionalGeneration",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "text_encoder",
+      "type_hint": [
+        "transformers",
+        "Gemma3ForConditionalGeneration"
+      ],
+      "variant": null
+    }
+  ],
+  "tokenizer": [
+    "transformers",
+    "GemmaTokenizer",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "tokenizer",
+      "type_hint": [
+        "transformers",
+        "GemmaTokenizer"
+      ],
+      "variant": null
+    }
+  ],
+  "transformer": [
+    "diffusers",
+    "EchoWMTransformer3DModel",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "transformer",
+      "type_hint": [
+        "diffusers",
+        "EchoWMTransformer3DModel"
+      ],
+      "variant": null
+    }
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLLTX2Video",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "vae",
+      "type_hint": [
+        "diffusers",
+        "AutoencoderKLLTX2Video"
+      ],
+      "variant": null
+    }
+  ],
+  "vocoder": [
+    "ltx2",
+    "LTX2VocoderWithBWE",
+    {
+      "pretrained_model_name_or_path": "Echo-Team/Echo-WM-Flash",
+      "revision": null,
+      "subfolder": "vocoder",
+      "type_hint": [
+        "ltx2",
+        "LTX2VocoderWithBWE"
+      ],
+      "variant": null
+    }
+  ]
+}

text_encoder/config.json

@@ -0,0 +1,124 @@
+{
+  "architectures": [
+    "Gemma3ForConditionalGeneration"
+  ],
+  "boi_token_index": 255999,
+  "dtype": "bfloat16",
+  "eoi_token_index": 256000,
+  "eos_token_id": [
+    1,
+    106
+  ],
+  "image_token_index": 262144,
+  "initializer_range": 0.02,
+  "mm_tokens_per_image": 256,
+  "model_type": "gemma3",
+  "text_config": {
+    "_sliding_window_pattern": 6,
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_logit_softcapping": null,
+    "bos_token_id": 2,
+    "cache_implementation": "hybrid",
+    "dtype": "bfloat16",
+    "eos_token_id": 1,
+    "final_logit_softcapping": null,
+    "head_dim": 256,
+    "hidden_activation": "gelu_pytorch_tanh",
+    "hidden_size": 3840,
+    "initializer_range": 0.02,
+    "intermediate_size": 15360,
+    "layer_types": [
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 131072,
+    "model_type": "gemma3_text",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 48,
+    "num_key_value_heads": 8,
+    "pad_token_id": 0,
+    "query_pre_attn_scalar": 256,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "full_attention": {
+        "factor": 8.0,
+        "rope_theta": 1000000,
+        "rope_type": "linear"
+      },
+      "sliding_attention": {
+        "rope_theta": 10000,
+        "rope_type": "default"
+      }
+    },
+    "sliding_window": 1024,
+    "sliding_window_pattern": 6,
+    "tie_word_embeddings": true,
+    "use_bidirectional_attention": false,
+    "use_cache": true,
+    "vocab_size": 262208
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.16.1",
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "image_size": 896,
+    "intermediate_size": 4304,
+    "layer_norm_eps": 1e-06,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 27,
+    "patch_size": 14,
+    "vision_use_head": false
+  }
+}

text_encoder/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "cache_implementation": "hybrid",
+  "do_sample": true,
+  "eos_token_id": [
+    1,
+    106
+  ],
+  "top_k": 64,
+  "top_p": 0.95,
+  "transformers_version": "5.16.1"
+}

text_encoder/model-00001-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:589aa3fa347646b2900edc2ef879caf7d5c85f9f5c1c1d46515913efc9e9cc72
+size 4939493376

text_encoder/model-00002-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d7508676d373ec6ddcc6cf79c34cea66aa16e2a93c5dead546fd85a07b1e9fb
+size 4931296608

text_encoder/model-00003-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f1d666658c3664f23a7cba92038ad9c82ca94e0b2447d09960b2ea393378a47
+size 4931296672

text_encoder/model-00004-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:727b51cc8ce5c7071bd1c4f4d801cb83b3320d9e0b38710dcb520bf7d6f66cc4
+size 4931296672

text_encoder/model-00005-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a1f0327a2d973743d76748abddacd10640a2e31ec7bf4fab5f025f02b056f2a
+size 4641409744

tokenizer/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

tokenizer/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daab2354f8a74e70d70b4d1f804939b68a8c9624dd06cb7858e52dd8970e9726
+size 33384567

tokenizer/tokenizer_config.json

@@ -0,0 +1,25 @@
+{
+  "backend": "tokenizers",
+  "boi_token": "<start_of_image>",
+  "bos_token": "<bos>",
+  "clean_up_tokenization_spaces": false,
+  "eoi_token": "<end_of_image>",
+  "eos_token": "<eos>",
+  "image_token": "<image_soft_token>",
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "model_specific_special_tokens": {
+    "boi_token": "<start_of_image>",
+    "eoi_token": "<end_of_image>",
+    "image_token": "<image_soft_token>"
+  },
+  "pad_token": "<pad>",
+  "processor_class": "Gemma3Processor",
+  "sp_model_kwargs": null,
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

transformer/config.json

@@ -0,0 +1,112 @@
+{
+  "_class_name": "EchoWMTransformer3DModel",
+  "_diffusers_version": "0.41.0.dev0",
+  "activation_fn": "gelu-approximate",
+  "attention_bias": true,
+  "attention_head_dim": 128,
+  "attention_out_bias": true,
+  "audio_attention_head_dim": 64,
+  "audio_cross_attention_dim": 2048,
+  "audio_cross_attn_mod": true,
+  "audio_ff_bias": true,
+  "audio_gated_attn": true,
+  "audio_hop_length": 160,
+  "audio_in_channels": 128,
+  "audio_num_attention_heads": 32,
+  "audio_out_channels": 128,
+  "audio_patch_size": 1,
+  "audio_patch_size_t": 1,
+  "audio_pos_embed_max_pos": 20,
+  "audio_sampling_rate": 16000,
+  "audio_scale_factor": 4,
+  "base_height": 2048,
+  "base_width": 2048,
+  "caption_channels": 3840,
+  "causal_offset": 1,
+  "cross_attention_dim": 4096,
+  "cross_attn_mod": true,
+  "cross_attn_timestep_scale_multiplier": 1000,
+  "ff_bias": true,
+  "gated_attn": true,
+  "in_channels": 128,
+  "norm_elementwise_affine": false,
+  "norm_eps": 1e-06,
+  "num_attention_heads": 32,
+  "num_layers": 48,
+  "out_channels": 128,
+  "patch_size": 1,
+  "patch_size_t": 1,
+  "perturbed_attn": true,
+  "pos_embed_max_pos": 20,
+  "qk_norm": "rms_norm_across_heads",
+  "rope_double_precision": true,
+  "rope_theta": 10000.0,
+  "rope_type": "split",
+  "timestep_scale_multiplier": 1000,
+  "ucpe_attention_dim": 1024,
+  "ucpe_block_indices": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29,
+    30,
+    31,
+    32,
+    33,
+    34,
+    35,
+    36,
+    37,
+    38,
+    39,
+    40,
+    41,
+    42,
+    43,
+    44,
+    45,
+    46,
+    47
+  ],
+  "ucpe_freq_base": 100.0,
+  "ucpe_freq_scale": 1.0,
+  "ucpe_image_height": 704,
+  "ucpe_image_width": 1280,
+  "ucpe_num_attention_heads": 8,
+  "ucpe_patches_x": 40,
+  "ucpe_patches_y": 22,
+  "use_keyframes_abs_pos_embedding": false,
+  "use_prompt_adaln_single": true,
+  "use_prompt_embeddings": false,
+  "vae_scale_factors": [
+    8,
+    32,
+    32
+  ]
+}

transformer/diffusion_pytorch_model-00001-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57a7d276f3df703811fa901862df1b3c6ff8858738dc9ef029353f70476f0692
+size 4989079504

transformer/diffusion_pytorch_model-00002-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7bbefe0d05b10a9e145e8f89663efc5bcdecd52088a4ee11d091e4db3ceb0b9
+size 4976466016

transformer/diffusion_pytorch_model-00003-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c06d77b224e2a0c68c4261c8ca459eceaa96d7571b674dc0658f2548c315dcc1
+size 4993239520

transformer/diffusion_pytorch_model-00004-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0505166a1858439025d7a2088a05bc051e62f14cf370c47acd5b67c67225ca8e
+size 4926109904

transformer/diffusion_pytorch_model-00005-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f36c255c1b6767f96093a16c82b0faff192972338a43e87b685de4730da8864
+size 4976037688

transformer/diffusion_pytorch_model-00006-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02af6d539ea5dbbec4b92ce814a564038e6e723bd0e77a915a2f9f216285f674
+size 4976013064

transformer/diffusion_pytorch_model-00007-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb7b15a1678cdc86c0e3ee91097bc56ab2584231abe005e888a177bfc17c05f7
+size 4976256400

transformer/diffusion_pytorch_model-00008-of-00008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6aedf50800eeb7dc4cf7b836e70931d5b68ca656a41ab733c763515f754f3476
+size 4776020920

vae/config.json

@@ -0,0 +1,93 @@
+{
+  "_class_name": "AutoencoderKLLTX2Video",
+  "_diffusers_version": "0.41.0.dev0",
+  "block_out_channels": [
+    256,
+    512,
+    1024,
+    1024
+  ],
+  "decoder_block_out_channels": [
+    256,
+    512,
+    512,
+    1024
+  ],
+  "decoder_causal": false,
+  "decoder_inject_noise": [
+    false,
+    false,
+    false,
+    false,
+    false
+  ],
+  "decoder_layers_per_block": [
+    4,
+    6,
+    4,
+    2,
+    2
+  ],
+  "decoder_spatial_padding_mode": "zeros",
+  "decoder_spatio_temporal_scaling": [
+    true,
+    true,
+    true,
+    true
+  ],
+  "down_block_types": [
+    "LTX2VideoDownBlock3D",
+    "LTX2VideoDownBlock3D",
+    "LTX2VideoDownBlock3D",
+    "LTX2VideoDownBlock3D"
+  ],
+  "downsample_type": [
+    "spatial",
+    "temporal",
+    "spatiotemporal",
+    "spatiotemporal"
+  ],
+  "encoder_causal": true,
+  "encoder_spatial_padding_mode": "zeros",
+  "in_channels": 3,
+  "latent_channels": 128,
+  "layers_per_block": [
+    4,
+    6,
+    4,
+    2,
+    2
+  ],
+  "out_channels": 3,
+  "patch_size": 4,
+  "patch_size_t": 1,
+  "resnet_norm_eps": 1e-06,
+  "scaling_factor": 1.0,
+  "spatial_compression_ratio": 32,
+  "spatio_temporal_scaling": [
+    true,
+    true,
+    true,
+    true
+  ],
+  "temporal_compression_ratio": 8,
+  "timestep_conditioning": false,
+  "upsample_factor": [
+    2,
+    2,
+    1,
+    2
+  ],
+  "upsample_residual": [
+    false,
+    false,
+    false,
+    false
+  ],
+  "upsample_type": [
+    "spatiotemporal",
+    "spatiotemporal",
+    "temporal",
+    "spatial"
+  ]
+}

vae/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:425f0dfa227dee5d0ff3d9720563370810409a439c302ca74f0f944057ce55c5
+size 1452233194

vocoder/config.json

@@ -0,0 +1,104 @@
+{
+  "_class_name": "LTX2VocoderWithBWE",
+  "_diffusers_version": "0.41.0.dev0",
+  "act_fn": "snakebeta",
+  "antialias": true,
+  "antialias_kernel_size": 12,
+  "antialias_ratio": 2,
+  "bwe_act_fn": "snakebeta",
+  "bwe_antialias": true,
+  "bwe_antialias_kernel_size": 12,
+  "bwe_antialias_ratio": 2,
+  "bwe_final_act_fn": null,
+  "bwe_final_bias": false,
+  "bwe_hidden_channels": 512,
+  "bwe_in_channels": 128,
+  "bwe_leaky_relu_negative_slope": 0.1,
+  "bwe_out_channels": 2,
+  "bwe_resnet_dilations": [
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ]
+  ],
+  "bwe_resnet_kernel_sizes": [
+    3,
+    7,
+    11
+  ],
+  "bwe_upsample_factors": [
+    6,
+    5,
+    2,
+    2,
+    2
+  ],
+  "bwe_upsample_kernel_sizes": [
+    12,
+    11,
+    4,
+    4,
+    4
+  ],
+  "filter_length": 512,
+  "final_act_fn": null,
+  "final_bias": false,
+  "hidden_channels": 1536,
+  "hop_length": 80,
+  "in_channels": 128,
+  "input_sampling_rate": 16000,
+  "leaky_relu_negative_slope": 0.1,
+  "num_mel_channels": 64,
+  "out_channels": 2,
+  "output_sampling_rate": 48000,
+  "resnet_dilations": [
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ]
+  ],
+  "resnet_kernel_sizes": [
+    3,
+    7,
+    11
+  ],
+  "upsample_factors": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "upsample_kernel_sizes": [
+    11,
+    4,
+    4,
+    4,
+    4,
+    4
+  ],
+  "window_length": 512
+}

vocoder/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c99dde145a982974bc4c39608b3b929f232f0126a55d8326a47ad54f486ae23c
+size 258305056