clarify licensing

README.md

@@ -21,10 +21,7 @@ Our testing has shown that the VAE is good at eliminating unwanted high-frequenc
 Note that the output is overall smoother and has significantly less artifacting around edges in high-detail regions.
 
 ## Licensing:
-This VAE is available under the terms of the [CC BY-NC-SA 4.0 Deed](https://creativecommons.org/licenses/by-nc-sa/4.0/).
-That means that you are free to use this model for personal, non-commercial use.
-You are also free to distribute this model alongside other (non-commercial) models, as long as you give credit.
-Please include the version number as well in case future models are released.
+This VAE is available under the terms of the [CC BY-NC-SA 4.0 Deed](https://creativecommons.org/licenses/by-nc-sa/4.0/).  This applies to the use of the model, deployment, and distribution of the model weights only. The license does not apply to images decoded by this VAE and you may release them under any license, even public domain, as long as you are not creating them for commercial purposes.  You are free and encouraged to distribute this VAE with models as long as you give credit and the VAE carries this license (the rest of the model does not need to share this license, although its distribution must be non-commercial), and I would ask that you include the version number so people can know if they need to get an updated version in the future.
 
 ## Training details:
 Overall training is fundamentally similar to LDM.  We used the same relative base weights for MAE, MSE, and LPIPS as used in LDM and in sd-vae-ft-mse in the case of LPIPS.  The discriminator's weight in the loss objective is dynamically set so that the gradient norm for the discriminator is half that of the reconstruction loss, just like LDM.  We used a similar discriminator to what LDM uses, except reparameterized to Wasserstein loss with a gradient penalty and with its group norm layers replaced with layer norms.