
NOTE: Although this might be an interesting Photography Lora, it doesn't look much like her photography style at all. 2.0 is currently training and already looks better on the first of 30 epochs. Once that version is released, I'll be re-naming this Lora accordingly and releasing 2.0 as a proper Natasha Braier photography Lora.
Update: V2.0 was trained and was a train-wreck so I archived it. Sorry, this will be the only version until I properly learn to do Style-based LoRAs.
Yet another update: V2.0 does work after-all, it just doesn't get prompted in a typical way. I'll upload 2.0 later after my own style Lora and try and explain.
This one is definitely under-trained. I suggest a strength above 1 (maybe 1.25-1.4) and maybe a higher guidance cfg. 2.0 will definitely be happening even if no one else wants it ;)
If you're not familiar with this photographer/cinematographer, her website is here: https://www.natashabraier.com/
Experiment and let me know if you get interesting results with this.
Trigger word may not be necessary, but try "braier_photo style"
描述:
v1.0, 2.0 will be trained very soon
训练词语: braier_photo style
名称: braier_photography_flux.safetensors
大小 (KB): 145971
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success