Add FP8 weight quantization guide to README

README.md

@@ -243,6 +243,38 @@ This moves each component (text encoder → transformer → VAE) to GPU only whe
 | `pipe.to("cuda")` | ~30 GB | Fastest | A100, H100, H200 |
 | `enable_model_cpu_offload()` | ~19 GB | Similar | RTX 4090, RTX 3090 |
 
+#### FP8 Weight Quantization (Optional)
+
+For further VRAM reduction, you can quantize the transformer weights to FP8 using [torchao](https://github.com/pytorch/ao):
+
+```bash
+pip install torchao
+```
+
+```python
+from torchao.quantization import quantize_, Float8WeightOnlyConfig
+
+pipe = DiffusionPipeline.from_pretrained(
+    "Motif-Technologies/Motif-Video-2B",
+    custom_pipeline="pipeline_motif_video",
+    trust_remote_code=True,
+    torch_dtype=torch.bfloat16,
+    guider=guider,  # see T2V example above
+)
+quantize_(pipe.transformer, Float8WeightOnlyConfig())
+pipe.enable_model_cpu_offload()
+
+output = pipe(prompt="...", height=736, width=1280, num_frames=121, num_inference_steps=50)
+export_to_video(output.frames[0], "output.mp4", fps=24)
+```
+
+This stores the transformer weights in FP8 (8-bit) instead of BF16 (16-bit), reducing peak VRAM from ~19 GB to ~15 GB while keeping all computation in BF16 precision.
+
+| Mode | Peak VRAM | Notes |
+|------|-----------|-------|
+| `enable_model_cpu_offload()` | ~19 GB | BF16 baseline |
+| `+ Float8WeightOnlyConfig` | ~15 GB | FP8 weights, BF16 compute |
+
 ### 🖥️ ComfyUI
 
 Official ComfyUI custom nodes for Motif-Video 2B are currently in development. Stay tuned for updates.