Trained for 333 epochs and 4000 steps.

Trained with datasets ['makima-text-embeds', 'makima-512']
Learning rate 0.0003, batch size 48, and 1 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: None

pytorch_lora_weights.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16ef819429b3dc1fbe038a746b70951a316e4b7a6ab4e78014d64b9cb15fef42
+oid sha256:e6b219c53e687d14f1e66b627b8099c3ecba14a1288aa591135a8788a2cd084a
 size 194456044