Update README.md

README.md

@@ -28,7 +28,10 @@ This model/pipeline is the product of my [LibreFlux ControlNet training repo](ht
 pip install -U diffusers==0.32.0
 pip install -U "transformers @ git+https://github.com/huggingface/transformers@e15687fffe5c9d20598a19aeab721ae0a7580f8a"
 ```
-
+Low VRAM:
+```py
+pip install optimum-quanto
+```
 # Load Pipeline
 ```py
 import torch
@@ -76,3 +79,44 @@ out = pipe(
         )
 out.images[0]
 ```
+# Load Pipeline ( Low VRAM )
+```py
+import torch
+from diffusers import DiffusionPipeline
+from optimum.quanto import freeze, quantize, qint8
+
+model_id = "neuralvfx/LibreFlux-ControlNet"  # <-- change if you used a different repo
+device = "cuda" if torch.cuda.is_available() else "cpu"
+dtype  = torch.bfloat16 if device == "cuda" else torch.float32
+
+pipe = DiffusionPipeline.from_pretrained(
+    model_id,
+    custom_pipeline=model_id,
+    trust_remote_code=True,       # required for your custom pipeline + modules
+    torch_dtype=dtype,
+    safety_checker=None           # optional; disable if not using a checker
+)
+
+quantize(
+    pipe.transformer,
+    weights=qint8,
+    exclude=[
+        "*.norm", "*.norm1", "*.norm2", "*.norm2_context",
+        "proj_out", "x_embedder", "norm_out", "context_embedder",
+    ],
+)
+
+quantize(
+    pipe.controlnet,
+    weights=qint8,
+    exclude=[
+        "*.norm", "*.norm1", "*.norm2", "*.norm2_context",
+        "proj_out", "x_embedder", "norm_out", "context_embedder",
+    ],
+)
+freeze(pipe.transformer)
+freeze(pipe.controlnet)
+
+pipe.enable_model_cpu_offload()
+
+```