Description
Issue Description
I'm currently experimenting with wildcards and used a wildcard list for a complete set of prompts. I used a batch count of 100 and a batch size of 4. When I checked some results I found that the metadata were inconsistent.
The prompts showing up in the image sets were all wrong for images 2, 3 and 4 of each set. There was always a new value taken from the wildcard file, while the image reflects correctly the prompt for the whole set. So only the first image in each set shows the correct prompt in the metadata. Not really a big thing, but a little inconvenient. ;-)
I use the integrated wildcard feature.
Version Platform Description
16:30:17-453589 INFO Starting SD.Next
16:30:17-455587 INFO Logger: file="F:\AI\SDNext\Basic\sdnext.log" level=INFO host="Fractal" size=20614086
mode=append
16:30:17-456591 INFO Python: version=3.11.11 platform=Windows bin="F:\AI\SDNext\Basic\venv\Scripts\python.exe"
venv="F:\AI\SDNext\Basic\venv"
16:30:17-595931 INFO Version: app=sd.next updated=2025-06-03 hash=894e15d3 branch=master
url=https://github.com/vladmandic/sdnext/tree/master ui=main
16:30:18-048545 INFO Version: app=sd.next latest=2025-06-03T15:14:48Z hash=894e15d3 branch=master
16:30:18-055545 INFO Platform: arch=AMD64 cpu=AMD64 Family 26 Model 68 Stepping 0, AuthenticAMD system=Windows
release=Windows-10-10.0.26100-SP0 python=3.11.11 locale=('de_DE', 'cp1252') docker=False
16:30:18-056544 INFO Args: ['--listen', '--listen']
16:30:18-063548 INFO CUDA: nVidia toolkit detected
16:30:18-138447 INFO Install: verifying requirements
16:30:18-207448 INFO Extensions: disabled=[]
16:30:18-208447 INFO Extensions: path="extensions-builtin" enabled=['Lora', 'sd-extension-chainner',
'sd-extension-system-info', 'sd-webui-agent-scheduler', 'sdnext-modernui',
'stable-diffusion-webui-rembg']
16:30:18-210447 INFO Extensions: path="extensions" enabled=[]
16:30:18-215447 INFO Startup: quick launch
16:30:18-216447 INFO Extensions: disabled=[]
16:30:18-216447 INFO Extensions: path="extensions-builtin" enabled=['Lora', 'sd-extension-chainner',
'sd-extension-system-info', 'sd-webui-agent-scheduler', 'sdnext-modernui',
'stable-diffusion-webui-rembg']
16:30:18-218450 INFO Extensions: path="extensions" enabled=[]
16:30:18-221447 INFO Installer time: total=1.14 latest=0.59 base=0.15 version=0.14 git=0.10
16:30:18-221447 INFO Command line args: ['--listen', '--listen'] listen=True args=[]
16:30:22-891927 INFO Torch: torch==2.7.0+cu128 torchvision==0.22.0+cu128
16:30:22-892927 INFO Packages: diffusers==0.34.0.dev0 transformers==4.52.3 accelerate==1.7.0 gradio==3.43.2
pydantic==1.10.21
16:30:23-317791 INFO Device detect: memory=31.8 default=balanced
16:30:23-323792 INFO Engine: backend=Backend.DIFFUSERS compute=cuda device=cuda attention="Scaled-Dot-Product"
mode=no_grad
16:30:23-363301 INFO Torch parameters: backend=cuda device=cuda config=FP16 dtype=torch.float16 context=no_grad
nohalf=False nohalfvae=False upcast=False deterministic=False tunable=[False, True] fp16=pass
bf16=pass optimization="Scaled-Dot-Product"
16:30:23-404303 INFO Device: device=NVIDIA GeForce RTX 5090 n=1 arch=sm_120 capability=(12, 0) cuda=12.8 cudnn=90701
driver=572.83
16:30:23-568323 INFO Available VAEs: path="models\VAE" items=0
16:30:23-569323 INFO Available UNets: path="models\UNET" items=0
16:30:23-570323 INFO Available TEs: path="models\Text-encoder" items=0
16:30:23-572322 INFO Available Models: safetensors="models\Stable-diffusion":3 diffusers="models\Diffusers":7
items=10 time=0.00
16:30:23-583322 INFO Available LoRAs: path="models\Lora" items=94 folders=3 time=0.01
16:30:23-590322 INFO Available Styles: path="models\styles" items=288 time=0.01
16:30:23-628325 INFO Available Detailer: path="models\yolo" items=10 downloaded=0
16:30:23-629327 INFO Load extensions
16:30:24-269573 INFO Extension: script='extensions-builtin\sd-webui-agent-scheduler\scripts\task_scheduler.py' Using
sqlite file: extensions-builtin\sd-webui-agent-scheduler\task_scheduler.sqlite3
16:30:24-278573 INFO Available Upscalers: items=71 downloaded=1 user=0 time=0.01 types=['None', 'Resize', 'Latent',
'AsymmetricVAE', 'DCC', 'VIPS', 'ChaiNNer', 'AuraSR', 'ESRGAN', 'RealESRGAN', 'SCUNet',
'Diffusion', 'SwinIR']
16:30:24-282573 INFO UI locale: name="en: English"
16:30:24-283576 INFO UI theme: type=Standard name="black-teal" available=13
16:30:25-655890 INFO Extension list is empty: refresh required
16:30:28-113154 INFO Local URL: http://localhost:7860/
16:30:28-115154 WARNING Public URL: enabled without authentication
16:30:28-115154 INFO External URL: http://192.168.2.7:7860
16:30:28-265704 INFO Public URL: http://87.154.15.245:7860
16:30:28-744269 INFO [AgentScheduler] Task queue is empty
16:30:28-745270 INFO [AgentScheduler] Registering APIs
16:30:28-829547 INFO Load model: select="Diffusers\black-forest-labs/FLUX.1-dev [0ef5fff789]"
16:30:29-131617 INFO HF login: token="hf_...BjuY" fn="C:\Users\volkm.cache\huggingface\token"
Progress 1.80s/it ██████████████ 100% 3/3 00:05 00:00 Loading checkpoint shards
Progress 1.03it/s ██████████████ 100% 2/2 00:01 00:00 Loading checkpoint shards
Progress 1.15s/it █████████ 100% 7/7 00:08 00:00 Loading pipeline components...
16:30:38-325238 INFO Offload: type=balanced op=init watermark=0.2-0.7 gpu=6.37-22.29:31.84 cpu=189.610 limit=0.00
16:30:38-348695 TRACE Offload: type=balanced op=skip gpu=1.820:1.820 perc=0.06 ram=34.490 current=cpu
dtype=torch.float16 quant=None module=FluxTransformer2DModel size=22.168
16:30:38-351696 TRACE Offload: type=balanced op=skip gpu=1.820:1.820 perc=0.06 ram=34.490 current=cpu
dtype=torch.float16 quant=None module=T5EncoderModel size=10.745
16:30:38-353697 TRACE Offload: type=balanced op=skip gpu=1.820:1.820 perc=0.06 ram=34.490 current=cpu
dtype=torch.float16 quant=None module=CLIPTextModel size=0.229
16:30:38-356696 TRACE Offload: type=balanced op=skip gpu=1.820:1.820 perc=0.06 ram=34.490 current=cpu
dtype=torch.float16 quant=None module=AutoencoderKL size=0.156
16:30:38-357696 TRACE Apply offload: time=0.03 type=balanced fn=load_diffuser:set_diffuser_offload
16:30:38-358696 INFO Model class=FluxPipeline modules=4 size=33.299
16:30:38-361695 INFO Load model: time=total=9.53 load=9.48 native=1024 memory={'ram': {'used': 34.49, 'total':
189.61}, 'job': 'Load', 'gpu': {'used': 1.82, 'total': 31.84}, 'active': 0.0, 'peak': 0.0,
'retries': 0, 'oom': 0, 'swap': 0}
16:30:38-364697 INFO Startup time: total=21.11 checkpoint=9.54 torch=4.48 launch=3.23 installer=0.97 libraries=0.76
extensions=0.64 gradio=0.62 ui-extensions=0.56 ui-models=0.42 api=0.40 diffusers=0.38
ui-networks=0.20 ui-txt2img=0.16 ui-control=0.16 app-started=0.16 ui-img2img=0.09
ui-defaults=0.07
16:30:38-404696 INFO UI: ready time=5.148
16:30:38-413697 INFO Browser session: user=None client=127.0.0.1 agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64;
rv:139.0) Gecko/20100101 Firefox/139.0
16:32:38-405521 TRACE Server: alive=True requests=54 memory=34.49/189.61 status='running' task='Load'
timestamp='20250605163028' current='' id='fad5956b4cc6421' job=0 jobs=0 total=1 step=0 steps=0
queued=0 uptime=135 elapsed=129.58 eta=None progress=0
Relevant log output
Backend
Diffusers
Compute
nVidia CUDA
UI
Standard
Branch
Master
Model
FLUX.1
Acknowledgements
- I have read the above and searched for existing issues
- I confirm that this is classified correctly and its not an extension issue