Run Locally
transparent-watercolor_Z_v2.safetensors
20 MB · SafeTensor
Downloads are served by Civitai. Some creators ask for a Civitai login first.
About This Model
ZIMAGE · LoRAs
by oosayam - transparent-watercolor(medium lora)
Creator notes
ZIT version
Two versions
Suitable weights are generally around 0.3–0.8
One is a Slider LORA trained purely from text
The other is trained using images generated by the Illustratious version of LORA
Actually, ZIT can produce watercolour paintings using the following prompt, but it shares the same issue as the slider model: inexplicably excessive mottled traces, overly fragmented details, and a lack of ethereal lightness. Hence both versions are uploaded for selection based on requirements.
Watercolour hues are light and translucent, with delicate layers of pale colour drifting like mist across the paper. The natural grain of the paper shines through, colour transitions are soft, edges dissolve naturally without any defined outlines, and expanses of colour spread evenly.
兩個版本
適合權重大概都是0.3-0.8左右
一個是純粹從文字訓練的slider lora
一個是用illustratious版本的LORA產生圖片訓練出來的
其實ZIT要出水彩畫用以下PROMPT就可以出來,但是和SLIDER一樣的問題,不知道為什麼斑駁痕跡太多,太細碎,不夠輕靈,所以兩個版本都上傳,根據需要使用吧
水彩顏色輕薄透明,淡彩層次如輕霧般覆蓋紙面,紙張紋理自然透出,色彩過渡柔和,邊緣自然消散,不帶任何線條輪廓,大片色彩均勻鋪展
Illustrious
Mixed by leco and image loras
suggest re-Forge use FreeU / SelfAttentionGuidance Integrated
and My UnstableInkDreamXL
and UnstableInkDreamXL_Neg
1.0 for sd 1.5
Merge of leco lora and landscape painting
weight 1~1.5
SHA256
D894291296E9F103C3603469610C6CA10862404838CC8309C354365D33941F5D
ModelTrends.ai Model ID
#23596
Flag content
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
Euler
Steps
9
Guidance
1
Seed
547322077088563
Size
1024x1024
Prompt
An anime-style watercolor painting showcases a handcrafted diorama, its horizontal frame revealing an exquisitely detailed miniature world brimming with wonder and childlike imagination. Entirely handmade from twigs, wood, clay, moss, dried flowers, and leaves, the diorama offers a rich palette of natural textures: the grain of raw wood, the softness of moss, and the texture of leaves. At the center of the frame stands a miniature cabin perched on an irregular network of branches. Its main structure is assembled and sculpted from thin, irregular twigs, wooden planks, and rustic clay. A small, simple balcony, surrounded by fluffy green moss and dotted with a few bouquets of small, dark pink dried flowers, extends from the cabin.
Seated on the balcony, an old gentleman made of clay and textured leaves, dressed in black jeans and a royal blue sweatshirt, with a black cap on his head, holds an open miniature book. The book's flat interior pages are illustrated with tiny, colorful plants and forest drawings. Nearby, a completely white German Shepherd, wearing a red fabric collar, sits with its head slightly tilted forward, its gaze fixed on the man's affectionate gaze.
The soft, natural light, diffused and of medium to low intensity, creates a peaceful atmosphere. This delicate light subtly highlights the open pages of the book. A faint halo of natural light filters through the openings of the surrounding miniature forest, delicately outlining the silhouettes of the man, the dog, and the treehouse. The color palette is dominated by the natural greens of the forest and earthy browns, while the wood and leaves provide soft, neutral tones. The background, densely covered with moss and dried flowers, is a deep, natural moss green. The image details are exceptionally sharp and clear, free from vignetting, halos, refraction, or grain, perfectly capturing the textures of the miniature materials. The overall atmosphere evokes a sense of peaceful camaraderie and natural storytelling, bearing the handcrafted mark and fairytale metaphor characteristic of artist Zho's ZH4O series.
Settings
Sampler
Heun
Steps
18
Guidance
3
Size
1212x1708
Prompt
i wont give the full prompt but you can start with this:
Seductive ginger Gen-Z model in golden medieval armor, ivy-covered ruins, epic siege, glowing hair, cinematic lighting, Cuphead-inspired, ultra-detailed. holding a banner crested with gems on the shield in fancy letters is written "To Buzz or not to BUZZ ?"
Settings
Sampler
Euler
Steps
9
Guidance
1
Seed
1197858245
Size
1152x1664
Prompt
Full-color drawing, sharp focus, cinematic framing, medium-distance shot, horizontal orientation, circular polarizer effect reducing water-surface glare and deepening sky and water color contrast, high-density precise linework combined with expressive line weight, clean cel shading with soft digital gradients, controlled cinematic lighting contrast.
A woman riding on a triggerfish. She has chestnut-colored hair that is thick, lush, and voluminous, worn loose around her head. She is wearing classical-inspired multicolored dress with simple, symmetrical lines and restrained detailing. air bubbles are floating around her.
The triggerfish is swimming forward just below the water surface, with a rigid body, visible dorsal and pectoral fins, and distinct patterned skin.
The environment shows underwater ocean with algae and coral reefs. Reflections on the water surface are visibly reduced by the polarizing effect.
Negative prompt
blurry, ugly, bad
Settings
Sampler
res_multistep
Steps
4
Guidance
1
Seed
4025700379
Size
1536x1184
Prompt
街景畫, transparent-watercolor, real world location
Negative prompt
worst_quality, bad_quality,bad_hands,very_displeasing,lowres,extra_digit,fewer_digits,jpeg ,watermark, instagram_username, twitter_username, patreon_username,frame,comic_fram
Settings
Sampler
Euler
Steps
8
Guidance
1.1
Seed
404214385
Size
832x1216
Weight used
0.5
Prompt
Create a whimsical watercolor painting featuring a beautiful woman standing outdoors in a college campus setting during autumn. She has shoulder-length feathered blonde hair, wearing a 1970s-style sweater and corduroy skirt with matching tights and a scarf. She holds a small satchel and a textbook, striking a thoughtful student pose as warm, nostalgic sunlight bathes the scene.
Settings
Sampler
res_multistep
Steps
4
Guidance
1
Seed
2638801390
Size
1536x1184
Prompt
街景畫, transparent-watercolor, real world location
Negative prompt
worst_quality, bad_quality,bad_hands,very_displeasing,lowres,extra_digit,fewer_digits,jpeg ,watermark, instagram_username, twitter_username, patreon_username,frame,comic_fram
Settings
Sampler
sa_solver
Steps
9
Guidance
1
Seed
2039495649
Size
1152x1664
Prompt
A full-color illustration with a painterly rendering style. The image is vertically oriented with centered framing. Brush strokes are broad and textured, with visible layering and uneven, broken edges. Shading is created through flat color contrast and overlapping paint layers rather than realistic lighting.
A black silhouette shaped like a woman is centered in the frame. it is ballet dancing. The silhouette is rendered as a continuous opaque black shape with no internal anatomical detail.
Clothing is indicated through overlaid painted shapes. A loose upper garment is drawn by a soft-edged shape in deep blue and purple tones covering the upper torso and arms. A skirt-like lower garment is drawn by a solid red shape around the hips and upper legs. The garment shapes have irregular boundaries and visible brush texture, sitting on top of the black silhouette without revealing underlying anatomy.
The environment is an abstract painted background filling the entire frame. Large vertical and horizontal blocks of yellow, emerald green, neon purple, and gray paint intersect behind the silhouette. The paint layers vary in opacity, with visible overlaps, rough edges, and scraped areas where underlying colors remain visible. No depth cues or realistic spatial indicators are present beyond paint overlap and color blocking.
Negative prompt
blurry, ugly, bad
Settings
Sampler
res_multistep
Steps
4
Guidance
1
Seed
2795253110
Size
1536x1184
Prompt
街景畫, transparent-watercolor, real world location
Negative prompt
worst_quality, bad_quality,bad_hands,very_displeasing,lowres,extra_digit,fewer_digits,jpeg ,watermark, instagram_username, twitter_username, patreon_username,frame,comic_fram
Similar Models
AI Models for Image, Video & Text Generation
transparent zit is a LoRA for the Zimage family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Zimage models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Text / LLM Models
411
Anima Models
509
Chroma Models
14
Flux Models
1,071
Flux 2 / Klein Models
1,237
MiniMax H3 Models
331
Hunyuan Models
197
Ideogram Models
20
Krea2 Models
674
Qwen Models
17
Qwen2 Models
36
Ltx2 Models
150
Stable Diffusion 1.5 Models
6,465
Stable Diffusion XL 1.0 Models
15,432
Zimage Models
804
Wan Models
919

Model
Trends
.ai
© ModelTrends.ai
|
Made in Japan
|
© 2026
