Update README.md

README.md

@@ -27,7 +27,9 @@ Please refer to [S-Lab License 1.0](https://github.com/IceClear/StableSR/blob/ma
 
 ### Limitations
 
-- TBD
+- StableSR still requires multiple steps for generating a large image beyond 512 or 768, which is much slower than GAN-based approaches.
+- StableSR sometimes cannot keep 100% fidelity due to its generative nature.
+- StableSR sometimes cannot generate perfect details under complex real-world scenarios.
 
 ### Bias
 While our model is based on a pre-trained Stable Diffusion model, currently we do not observe obvious bias in generated results.
@@ -55,8 +57,9 @@ StableSR is an image super-resolution model finetuned on [Stable Diffusion](http
 
 We currently provide the following checkpoints:
 
-- [stablesr_000117.ckpt](https://huggingface.co/Iceclear/StableSR/resolve/main/stablesr_000117.ckpt): Diffusion model finetuned on DF2K_OST dataset for 117 epochs.
+- [stablesr_000117.ckpt](https://huggingface.co/Iceclear/StableSR/resolve/main/stablesr_000117.ckpt): Diffusion model finetuned on [SD2.1-512base](https://huggingface.co/stabilityai/stable-diffusion-2-1-base) with DF2K_OST dataset for 117 epochs.
 - [vqgan_cfw_00011.ckpt](https://huggingface.co/Iceclear/StableSR/resolve/main/vqgan_cfw_00011.ckpt): CFW module with fixed VQGAN trained on synthetic paired data for 11 epochs.
+- [stablesr_768v_000139.ckpt](https://huggingface.co/Iceclear/StableSR/blob/main/stablesr_768v_000139.ckpt): Diffusion model finetuned on [SD2.1-768v](https://huggingface.co/stabilityai/stable-diffusion-2-1) with DF2K_OST dataset for 139 epochs.
 
 ## Evaluation Results 
 See [Paper](https://arxiv.org/abs/2305.07015) for details.