Add precomputed AdaLN tables for the 8-step schedule (skip the 26 GB projections)

README.md

@@ -15,5 +15,20 @@ RTX PRO 6000) or a DGX Spark.
 - **VAE:** LynnReal lightweight video VAE with Kijai's INT8 weights; H3 audio VAE.
 - **Sampling:** `fastvideo_inference.json` holds the 8-step DMD schedule, which FastVideo reads automatically.
 
+## Machines with less than 64 GB of system RAM
+
+The 50 AdaLN timestep projections are 26 GB in BF16. `transformer/adaln_tables.pt` (155 MB) holds their precomputed
+outputs for the 8-step schedule, so FastVideo can skip loading them:
+
+```python
+import os
+from huggingface_hub import hf_hub_download
+
+os.environ["FASTVIDEO_H3_ADALN_TABLE"] = hf_hub_download(
+    "FastVideo/FastVideo-FastH3-8-Step-V2-NVFP4-Consumer", "transformer/adaln_tables.pt")
+```
+
+The table covers only the shipped schedule (`fastvideo_inference.json`); FastVideo raises an error for other timesteps.
+
 For multi-GPU data-center serving, use [FastVideo/FastVideo-FastH3-8-Step-V2-NVFP4](https://huggingface.co/FastVideo/FastVideo-FastH3-8-Step-V2-NVFP4).
 Smaller, faster experimental variant: [FastVideo/FastVideo-FastH3-Trim-8-Step-NVFP4](https://huggingface.co/FastVideo/FastVideo-FastH3-Trim-8-Step-NVFP4).

transformer/adaln_tables.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ae3bdebed7d92a71ba930da9cabacafed92c3c2aa550a284f324aba7841a234
+size 154971509