<pre><code>These are **photographic grade / light LoRAs for Krea 2**, trained on my own Canon raw files from one trip each,
developed to a specific look before training. They are not subject or character LoRAs and they need **no trigger word**:
the look is applied to whatever you prompt. Every version was trained on exactly the same photos as its sibling, only the
grade changed, so the two versions of each model are true opposites you can A/B on one seed.

**Base and settings.** Trained with ai-toolkit on Krea-2-Raw (LoRA rank 32 / alpha 32, lr 1e-4, batch 1, buckets 512/768/1024,
2,500 steps, flow-matching). Tested on **Krea 2 Turbo FP8** in ComfyUI: 8 steps, CFG 1, `er_sde` / `simple`, 1216×832 and 832×1216,
text encoder `qwen3vl_4b_fp8_scaled`. Load with a normal LoRA loader on both model and CLIP.

**Strength.** 1.0 is the trained look. 0.5 to 0.7 gives a half-strength grade that still reads as the look. Above 1.2 the
grade gets heavier and compositions start to drift.

**Prompting.** Plain photographic prompts work best: `a photograph of …` or `a cinematic photograph, …`. No trigger word;
in fact the token from an earlier build tended to leak into the image as literal text, so these were retrained without one.

**img2img.** The Moab and Kauai versions work as colour grades on your own photos at denoise 0.45–0.55 (structure kept,
grade applied). The Olympic Peninsula versions are place-and-light LoRAs rather than grades and barely change a sunny
source below denoise 0.65; see their notes.

**What they do to off-topic prompts.** All six leave the composition of unrelated subjects alone (portraits, food, city
streets, wildlife) and only shift tone, with one exception noted on the Olympic Peninsula page: those two override the
prompt's lighting toward overcast daylight.

All showcase images are one seed rendered twice, LoRA on (top) and LoRA off (bottom), same prompt, so what you see
is exactly what the LoRA adds.</code></pre><p></p>

Slickrock Light

<pre><code>**The quiet one.** Pale, matte, lifted blacks, softened whites, restrained colour and a neutral hue: the black point
rises from about 2 to 7 (Lab L) and saturation drops slightly, brightness and hue barely move. Of all six it changes
composition the least; it changes tone and nothing else, on every subject, in txt2img and img2img alike.

- **Best for:** nostalgic and archival looks: found-photograph sci-fi (a 1970s Mars colony), storybook low fantasy,
  anything that should feel like a print rather than a screen. Also the safest general-purpose "take the digital edge
  off" LoRA of the set.
- **Keeps the prompt's light:** yes, fully.
- **Quirks:** it is subtle at 1.0, so on already-soft prompts it can read as merely matte. Push to 1.2–1.3 if you want it
  to announce itself. Shares the Mesa Arch memorisation with Amber.
- **Strength:** 1.0–1.3.
- **img2img:** the cleanest of the six as a pure grade at denoise 0.5.

**Pair them:** Amber and Faded Film are exact opposites on one seed: crushed warm chrome vs lifted pale print.
</code></pre>

Files with SHA256: 1d83c60309ab59820de8083d21925a445ce78a3bde581880bd0a596b0741ccf2