# V2 This model was finetuned from Stable Diffusion's 1.4 release with batch size 3 and `5e-6` learning rate on a single Radeon Pro W6800 GPU. Captions use space separated tags. Spaces within tags should be replaced with underscores. The autoencoder was finetuned for ~120,000 steps before finetuning the rest of the model. ## Full finetune release * [`sd1.4 finetuned v2 derpibooru gs=120000.ckpt`](https://mega.nz/file/FdJn3DTY#2AeH-JxJV7i3YJGoE4j2maiX2JzLNmrMRcsPYehSLXA) * [`sd1.4 finetuned v2 derpibooru gs=210000.ckpt`](https://mega.nz/file/ZJQVVAIS#WZCiS32lBmYxRdNjwIaPWjspJk9Y8eyTzalb-oj55XQ) Global step count in filename doesn't match in the state_dict due to the way I resumed at one point. TODO: Sample images, release more checkpoints later. ## Short finetune / Beta (2022-10-11) Model This checkpoint was captured after ~26,000 steps (~80,000 images) of training, which took 12 hours. This isn't even a full epoch.
![](./v2-beta-glim1.png)
pony s5_starlight unicorn glimmer this_will_end_in_communism scr100 bust mare portrait by=pierogarts female safe looking_at_you solo
This shows that finetuning to a reasonable quality level does not need a huge cluster of expensive datacenter GPUs. * [`sd1.4 finetuned v2 derpibooru e=0000 gs=026400.ckpt`](https://mega.nz/file/dFYmUQDS#8AZCBCA8btOrPZCVhJ1hWnL1NGzJCzQV3hvTGPOCKpw) # V1 (2022-10-01) Model This model was finetuned for ~500,000 steps at multiple different batch sizes and learning rates. It performs better than the current V2 model due to its longer training time, but I expect it to be surpassed soon. The training captions used comma separated tags.
![](./v1-fluttershy1.png)
fluttershy, blush, pony, pomf, cute, adorable, daww, solo, scenery, scr800, flying, pegasus
* [`sd1.4 finetuned derpibooru safe suggestive e=0002 gs=287000.ckpt`](https://mega.nz/file/MMoHwRCa#p06t5SVCSSlEvwVmDOf9B8StDw7BJ0gidS7Y62HjDtM) * [`sd1.4 finetuned derpibooru safe suggestive e=0007 gs=460000.ckpt`](https://mega.nz/file/UJx0TDZS#i6Y12ajvndy0YKbM6pvOTQuQYI5zpKeDTl73l9S8XsI) # Training Code These models were trained using [github:LunNova/translunar-diffusion](https://github.com/LunNova/translunar-diffusion). Please see [the stable diffusion training notes](/articles/stable-diffusion-training-notes/index.md) for more details. # Model Licenses These releases are all licensed under the [CreativeML Open RAIL-M License](https://github.com/CompVis/stable-diffusion/blob/main/LICENSE), as required for derivatives of the original stable diffusion model.