The monotonic-training counterpart to alvdansen/illustration-1.0-qwen-image. Same 244-image dataset, same hyperparameters, same hardware — trained without the chained dataset rotation. Published alongside the chained version so the paired comparison in Forgetting on Purpose is reproducible. Read the paper · Source on GitHub.

For general use, the chained version is the better pick. This one is here for anyone who wants to run their own comparison.

Usage

No trigger word. Same inference recipe as the chained twin:

Sampler: euler
Scheduler: simple
CFG: 3.5
Steps: 45
LoRA strength: 0.8–1.0

Training Details

Base model: Qwen-Image (FP8 quantized, text encoder FP8)
Training steps: 59,000, monotonic full-dataset training from step zero
Rank/Alpha: 42/42
Learning rate: 5e-5
Optimizer: AdamW 8-bit
Caption dropout: 0.35
EMA: enabled (decay 0.99)
Noise scheduler: flowmatch
Precision: bf16 with qfloat8 quantization
Dataset: same 244-image combined dataset as the chained twin
Trainer: ai-toolkit by Ostris
Hardware: NVIDIA RTX 6000 Ada (A6000, 48 GB VRAM)

Identical to the chained twin in every parameter except the dataset schedule.

alvdansen/illustration-1.0-qwen-image-baseline

Illustration 1.0 — Qwen-Image (Baseline)

Usage

Training Details

alvdansen/
illustration-1.0-qwen-image-baseline