Update README.md

README.md

@@ -16,7 +16,7 @@ This model is a **26.4M** LoRA distilled version of SD1.5 model that is able to
 
 
 <p align="center">
-   <img style="width:700px;" src="images/hf_grid.png">
+   <img style="width:700px;" src="images/flash_sd.jpg">
 </p>
 
 # How to use?
@@ -53,7 +53,7 @@ image = pipe(prompt, num_inference_steps=4, guidance_scale=0).images[0]
 </p>
 
 # Training Details
-The model was trained for 20k iterations on 2 H100 GPUs (representing approx. **26 hours** of training). Please refer to the [paper]() for further parameters details. 
+The model was trained for 20k iterations on 2 H100 GPUs (representing approx. a total **26 GPU hours** of training). Please refer to the [paper]() for further parameters details. 
 
 **Metrics on COCO 2017 validation set (Table 1)**
 - FID-5k: 22.6 (2 NFE) / 22.5 (4 NFE)