Update README.md

README.md

@@ -37,8 +37,8 @@ This card in particular describes `Photoroom/prx-512-t2i-dc-ae-sft-distilled`, o
 - **Architecture:** PRX (MMDiT-like diffusion transformer variant)  
 - **Latent backbone:** [DC-AE VAE](https://arxiv.org/abs/2410.10733)  
 - **Text encoder:** T5-Gemma-2B-2B-UL2
-- **Training stage:** Supervised fine-tuning (SFT)
-- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0) and 8-step distilled
+- **Training stage:** Supervised fine-tuning (SFT) and 8-step distilled
+- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
 
 For other checkpoints, browse the full [PRX collection](https://huggingface.co/collections/Photoroom/prx).
 
@@ -55,7 +55,7 @@ pipe = PRXPipeline.from_pretrained(
 ).to("cuda")
 
 prompt = "A front-facing portrait of a lion in the golden savanna at sunset"
-image = pipe(prompt, num_inference_steps=28, guidance_scale=5.0).images[0]
+image = pipe(prompt, num_inference_steps=8, guidance_scale=5.0).images[0]
 image.save("lion.png")
 ```