Qwen is evolving very fast.

Tungsten has to keep running the race.

That's why I'm curious to see if this lora works on the simple Qwen image source. Luckily, we have the realistic workflow thing.

I'm not the author, here is what he writes in Civ:

"This LoRA, built on Qwen-Edit 2511, incrementally maps images of any visual style into the photographic domain. The training objective is to minimize perceptual distance rather than to perform pixel-level reconstruction.

Training Protocol

Dataset

7 style-paired subsets, 200 source images ≥ 1024 px, covering portraits, landscapes, and objects.

Training Pipeline

Two-stage tuning: (a) trigger-word + long-caption pairs, (b) trigger-word only.

Layer-wise merge with Anything2Real α; weights manually tuned to preserve low-frequency structure while replacing high-frequency texture.

Recommended Positive Prompt

transform the image to realistic photograph. {detailed description}

Note: {detailed description} is optional. Style-related tokens such as “anime”, “cyber”, or “oil-painting” are strictly forbidden; their inclusion will inject out-of-domain noise.

Strength Tuning

Default range: 0.8–1.0.

If over-smoothing or texture loss occurs, decrease in 0.1 steps down to 0.5.

Should realism remain insufficient, blend with an additional photorealistic LoRA and adjust to taste.

Because “realism” is inherently subjective, first modulate strength or switch base models rather than further increasing the LoRA weight."