Trained for 3 epochs and 1600 steps.

Trained with datasets ['text-embeds', 'q1sV16-1024-crop', 'q1sV16-512-crop', 'gen-and-inpaint-q1s-1024-reg']
Learning rate 6e-05, batch size 1, and 3 gradient accumulation steps.
Used DDPM noise scheduler for training with epsilon prediction type and rescaled_betas_zero_snr=False
Using 'trailing' timestep spacing.
Base model: black-forest-labs/FLUX.1-dev
VAE: black-forest-labs/FLUX.1-dev

pytorch_lora_weights.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9c6eab0e9c7827e63e410536fb0aff212e074ed23b1951af3af13e09a4a714d2
+oid sha256:df381877707b4810774b360892f05509f789f71e24b654d37efdefe8702fd7f7
 size 109535132