1.0.5 — per-model demo subjects, and a new FLUX style · cec7f9f9cd - StyleRef-styleref-comfyui

1.0.5 — per-model demo subjects, and a new FLUX style

The two template sets shared one GRID_SUBJECTS list even though they load
different styles, so the "fully dressed"/"dressed" phrasing was leaking into
the FLUX set. That phrasing exists for Z-Image alone: the Renaissance style
compiles an empty negative and Z-Image has no negative channel to hold one.

Split it into ZIMAGE_GRID_SUBJECTS and FLUX_GRID_SUBJECTS, each with its own
demo subject. Z-Image's three are unchanged. FLUX gets three built around
props, spill and cast shadow.

The FLUX set also moves from Noir Low-Key Portrait to Raw Theatrical
Cinematic (azch4ptx-2ca7d3a07f37) — same reasoning as before (light over
brushwork, subject-agnostic, carries inspiration images), a style whose
staged-clutter identity suits the new subjects.

Known gap: the five FLUX template thumbnails still show the old style and
subjects. They need re-rendering in a follow-up.

This commit is contained in:

Behzad

2026-07-30 00:55:33 +02:00

parent 4862b286a6

commit cec7f9f9cd

10 changed files with 72 additions and 47 deletions

									
										styleref_api.py
									
		+1
		-1
	
												View File
												
				@@ -25,7 +25,7 @@ SITE_URL = "https://styleref.io"

				# Identifies this surface in StyleRef's API telemetry (recordApiCall reads it).

				CLIENT_TAG = "comfyui"

				USER_AGENT = "styleref-comfyui/1.0.4 (+https://styleref.io)"

				USER_AGENT = "styleref-comfyui/1.0.5 (+https://styleref.io)"

				DEFAULT_TIMEOUT_S = 30