Trained for 0 epochs and 30000 steps.

Trained with datasets ['assistant-offline-text', 'assistant-real-v3-cc12m-512', 'assistant-real-v3-cc12m-1024', 'assistant-real-v3-cc12m-1536', 'assistant-real-v3-cc12m-2048', 'assistant-real-v3-e621-512', 'assistant-real-v3-e621-1024', 'assistant-real-v3-e621-1536', 'assistant-real-v3-e621-2048']
Learning rate 0.0001 and batch size 2.
Training objective: flow_map.
Base model: black-forest-labs/flux-3-image-klein
VAE: black-forest-labs/flux-3-image-klein

pytorch_lora_weights.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a22385b2f1e48b9ebc2269f1c5748cedc73a68086b9e3707cd69be0056f52be
-size 135314984
+oid sha256:a63989cf0ea6f365de9da9d73378709f374ac51f711346246c334278088d254d
+size 135312896