Add accurate Fast model index and complete-package download instructions

README.md

@@ -14,7 +14,13 @@ Self-contained weights for WorldCrafter Fast image-to-video, text-to-video, and
 
 The two transformer branches have independent DMD LoRA and camera adapters. The RepEncoder, text encoder, tokenizer, VAE, and scheduler are included here. No WorldCrafter-Base files are needed for Fast inference.
 
-From the code repository root:
+From the code repository root, download the complete package, including its model index and both transformer branches:
+
+```bash
+hf download TencentARC/WorldCrafter-Fast --local-dir weights/WorldCrafter-Fast
+```
+
+Then run inference:
 
 ```bash
 python inference.py --model-type fast --model-path weights/WorldCrafter-Fast --output-path outputs/fast.mp4
@@ -22,6 +28,8 @@ python inference.py --model-type fast --model-path weights/WorldCrafter-Fast --o
 
 Image-to-video uses five high-noise steps and one low-noise step per chunk. Text-to-video uses four high-noise and eight low-noise steps for the first chunk, then two high-noise and four low-noise steps. Compilation is optional with `--enable-compile`.
 
+`model_index.json` describes the two transformer branches, their adapters, and the shared components. Loading uses `worldcrafter.WorldCrafter.from_pretrained(..., model_type="fast")` through the WorldCrafter code; this package is not a standalone `diffusers.DiffusionPipeline.from_pretrained` checkpoint.
+
 `inference_config.json` resolves shared components from this directory (`shared_components: "."`). Keep all component folders together. `manifest.json` records checkpoint provenance and validation hashes; `SHA256SUMS` covers the packaged files.
 
 ## Paper and Resources

SHA256SUMS

@@ -1,4 +1,4 @@
-747abfc4409f58a0c7ac77d8518f97467f99f7a409dd9a40da727b2d57d34989  README.md
+6c86d51bacc2a12ce1d60e6d3a05c2785559dcb78b00ab6cbddc7f19f01bd22e  README.md
 f1577901740c5b77d76cfb4732fdf4dbd8afc87e6232ecf5db6e1209594bfb59  adapter_high_noise/dmd_timestep_contract.json
 e46e41944879e09f66f5572b13f29b562ddc96a44ebfef59804357d022d765df  adapter_high_noise/pytorch_lora_weights.safetensors
 333dba388746b475114cd664d6fff498b26d7eef4d366f882f4cb7c26077c39e  adapter_high_noise/repencoder_frozen.json
@@ -9,6 +9,7 @@ a0f2884a8e873ea3db316ccd1a0502c98e1a335b52c277aecdbbbcfe1212dc36  adapter_low_no
 72e4a61fe7c6fb1af9285a4f09f545919f3b625c69647ebfe7565a6bf04934cc  adapter_low_noise/transformer_partial.pth
 b25c5764a00ead6b909d24920ae289c69c0e97494baf089a4778e0a8c180abb9  inference_config.json
 894fec880301593ab7f9727647041e5fa07d401fddbd66979993fad131adb1da  manifest.json
+dad03f7d77639589e7ece235717d1b824214064f742c5a82f019b2162d2d4030  model_index.json
 bf2dbb002da475cf00ce4081369c1bc60a7199d8c5a5e92d801d8670e396d153  repencoder/config.json
 6b2b0ff728631b83675847f813cb8b834f62ac3d8c0ad9bf774fbd2077669236  repencoder/manifest.json
 5e86abf6f1ecf86f52610be115827d123113fdfa1aab00630d76a2d2a1e938fc  repencoder/model.safetensors

model_index.json

@@ -0,0 +1,62 @@
+{
+  "_class_name": "WorldCrafter",
+  "_module": "worldcrafter",
+  "_format": "worldcrafter_fast_model_index_v1",
+  "_diffusers_version": "0.37.0",
+  "_loading": {
+    "entrypoint": "worldcrafter.WorldCrafter.from_pretrained",
+    "model_type": "fast",
+    "requires_custom_loader": true,
+    "inference_config": "inference_config.json",
+    "manifest": "manifest.json"
+  },
+  "_pipeline": {
+    "library": "worldcrafter",
+    "class_name": "WorldCrafterPipeline",
+    "is_distilled": true
+  },
+  "transformer_high_noise": [
+    "worldcrafter",
+    "WorldCrafterTransformer3DModel"
+  ],
+  "transformer_low_noise": [
+    "worldcrafter",
+    "WorldCrafterTransformer3DModel"
+  ],
+  "repencoder": [
+    "worldcrafter",
+    "RepEncoder"
+  ],
+  "text_encoder": [
+    "transformers",
+    "UMT5EncoderModel"
+  ],
+  "tokenizer": [
+    "transformers",
+    "T5TokenizerFast"
+  ],
+  "vae": [
+    "diffusers",
+    "AutoencoderKLWan"
+  ],
+  "scheduler": [
+    "worldcrafter",
+    "WorldCrafterScheduler"
+  ],
+  "_adapters": {
+    "high_noise": {
+      "transformer": "transformer_high_noise",
+      "lora_weights": "adapter_high_noise/pytorch_lora_weights.safetensors",
+      "camera_weights": "adapter_high_noise/transformer_partial.pth",
+      "timestep_contract": "adapter_high_noise/dmd_timestep_contract.json",
+      "repencoder_reference": "adapter_high_noise/repencoder_frozen.json"
+    },
+    "low_noise": {
+      "transformer": "transformer_low_noise",
+      "lora_weights": "adapter_low_noise/pytorch_lora_weights.safetensors",
+      "camera_weights": "adapter_low_noise/transformer_partial.pth",
+      "timestep_contract": "adapter_low_noise/dmd_timestep_contract.json",
+      "repencoder_reference": "adapter_low_noise/repencoder_frozen.json"
+    }
+  }
+}