TSNAIMix aims to be a 2d focused model with good stability, inteligence and versatility
You can try online generation of my other models Here
Recommended Settings:
Sampling Method: Euler A
Scheduler: SGM Uniform
CFG Scale: 6-9 (7 works better in general)
Steps: 25
RESOLUTIONS:
1536 x 640
1344 x 768
1216 x 832
1152 x 896
1024 x 1024
896 x 1152
832 x 1216
768 x 1344
640 x 1536
This is a Mix of ChenkinNoobXL+TSRMix with the objective of achieve the stability and inteligence of TSRMix + updated booru knowledge provided by Chenkin
Model License
This model inherits the same license from ChenkinNoobXL:
Chenkin Noob XL （CKXL ） - v0.1 | NoobAI Checkpoint | Civitai
Description
FAQ
Comments (8)
Very cinematic/filmic, great composition, works super well with style LoRAs. Really happy with the lineweight. Lighting and reflections are solid, super atmospheric. Handles gradient backgrounds and ethereal/abstract prompts really well.
Only minor downsides based on my workflow/techniques (YMMV): color contrast can be a bit low (easily fixed in postprocessing, not a dealbreaker), and Sometimes too much film grain in dark nighttime/shadow shots.
Love it :), i had been using TSRMIX5 as my go to for a while now, will probably switch to this checkpoint as it allows me to be much closer to my desired arstyle.
Thank you for the feedback! I'll try to fix the imperfections you mentioned. I'm glad that you like it!
@ZaeY I played around with it more, and its 100% because of my workflow, The style Lora and style tags im using are quire de-saturated, i hadn't noticed before because TSRMIX5 masked it. Time for me to tweak my workflow.
Nice, with Chenkin NoobAI's dataset! So far it's pretty work well with artist tags and style loras. ^^
Finally Noob Ver! Thank you so much!
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.