Create README.md

README.md

@@ -0,0 +1,34 @@
+# flux_quantized_half
+
+## Model Overview
+
+`Silan10/flux_quantized_half` is a **half-precision (FP16) variant** of the
+[`black-forest-labs/FLUX.1-dev`](https://huggingface.co/black-forest-labs/FLUX.1-dev)
+text-to-image model.
+
+This repository only changes the **numerical precision of the weights** to
+`torch.float16` using PyTorch.
+
+- **Base model:** `black-forest-labs/FLUX.1-dev`
+- **Precision:** FP16 (`torch.float16`)
+
+## Usage (Diffusers)
+
+```python
+import torch
+from diffusers import FluxPipeline
+
+pipe = FluxPipeline.from_pretrained(
+    "Silan10/flux_quantized_half"
+)
+pipe.to("cuda")  # or pipe.enable_model_cpu_offload() for low VRAM
+
+prompt = "A detailed concept art of a futuristic city at sunset"
+image = pipe(
+    prompt,
+    guidance_scale=3.5,
+    num_inference_steps=20,
+    max_sequence_length=256,
+).images[0]
+
+image.save("flux_half_sample.png")