Use Qwen Image 2.1 for refiner base

This commit is contained in:
Morpheus
2026-10-28 20:48:02 +00:00
parent 5668acbdea
commit 7ffd45539e
6 changed files with 76 additions and 15 deletions
+3 -2
View File
@@ -8,14 +8,15 @@ models.
## Current test
The first test uses the same local Qwen NoctQ v4 → FLUX.3 Klein pipeline as the
initial refiner experiment, but with the new reference image
The first test used Qwen NoctQ v4. The follow-up uses the proven Qwen Image 2.1
I2I stack and settings from `optimized/I2I/qwen/`, with the new reference image
`refiner_reference.jpg`. The FLUX stage receives the Qwen pixels and has no
text prompt, so it cannot replace the subject from text conditioning.
| ID | Input | Prompt | Base stage | Refiner stage | Resolution | Generation time (s) | Status | Review |
|---|---|---|---|---|---:|---:|---|---|
| QRF-I2I-01 | `refiner_reference.jpg` | Canonical shared I2I prompt | NoctQ v4, 8 steps, CFG 1, Euler ancestral / beta | FLUX.3 Klein, 8 steps, guidance 1, Euler ancestral / simple, denoise 0.11 | 1088×1920 | 115.720 uncached; 0.274 cached | Completed | Base and refined outputs are nearly identical at denoise 0.11; [base PNG](results/QRF-I2I-01-base.png) vs [refined PNG](results/QRF-I2I-01.png) |
| QRF-I2I-02 | `refiner_reference.jpg` | Canonical shared I2I prompt | Qwen Image 2.1, 50 steps, CFG 3.5, Euler CFG++ / beta | FLUX.3 Klein, 8 steps, guidance 1, Euler ancestral / simple, denoise 0.11 | Base 1080×1920; refined 1072×1920 | 145.129 | Completed | More realistic base; [base PNG](results/QRF-I2I-02-base.png) vs [refined PNG](results/QRF-I2I-02.png) |
## Planned follow-ups