First model version

93x720p/config.json

@@ -0,0 +1,47 @@
+{
+  "_class_name": "OpenSoraT2V",
+  "_diffusers_version": "0.28.0",
+  "activation_fn": "gelu-approximate",
+  "attention_bias": true,
+  "attention_head_dim": 96,
+  "attention_mode": "xformers",
+  "attention_type": "default",
+  "caption_channels": 4096,
+  "cross_attention_dim": 2304,
+  "decay": 0.999,
+  "double_self_attention": false,
+  "downsampler": null,
+  "dropout": 0.0,
+  "in_channels": 4,
+  "interpolation_scale_h": 1.5,
+  "interpolation_scale_t": 1.0,
+  "interpolation_scale_w": 2.0,
+  "inv_gamma": 1.0,
+  "min_decay": 0.0,
+  "norm_elementwise_affine": false,
+  "norm_eps": 1e-06,
+  "norm_num_groups": 32,
+  "norm_type": "ada_norm_single",
+  "num_attention_heads": 24,
+  "num_embeds_ada_norm": 1000,
+  "num_layers": 32,
+  "num_vector_embeds": null,
+  "only_cross_attention": false,
+  "optimization_step": 400,
+  "out_channels": 4,
+  "patch_size": 2,
+  "patch_size_t": 1,
+  "power": 0.6666666666666666,
+  "sample_size": [
+    90,
+    160
+  ],
+  "sample_size_t": 24,
+  "upcast_attention": false,
+  "update_after_step": 0,
+  "use_additional_conditions": null,
+  "use_ema_warmup": false,
+  "use_linear_projection": false,
+  "use_rope": true,
+  "use_stable_fp32": true
+}

93x720p/diffusion_pytorch_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7eb4e905781488a04995c85c4f1237fb5d0fb4ffae77a49692a9ca65ee1c8a8a
+size 11087709680