Flux Lora Rtmi
模型简介
该模型是一个基于扩散模型的LoRA适配器,专门用于生成具有《重返猴岛》游戏风格的图像。通过文本提示词触发,能够生成卡通化、夸张风格的场景和角色设计。
模型特点
RTMI风格生成
专门生成具有《重返猴岛》游戏独特卡通风格的图像
LoRA适配器
作为轻量级适配器,可在基础模型上灵活调整生成风格
高分辨率输出
支持1024x1024高分辨率图像生成
模型能力
文本生成图像
风格化图像生成
角色设计
场景设计
使用案例
游戏设计
游戏角色设计
生成具有RTMI风格的游戏角色形象
如示例中的程序员盖布拉什角色
游戏场景设计
创建夸张卡通风格的场景
如示例中的电视巢穴和海滩场景
概念艺术
概念插画
快速生成风格化概念艺术
如示例中的悬崖和餐厅场景
🚀 Flux Lora Rtmi
这是一个基于 《重返猴岛》(Return to Monkey Island) 游戏截图训练的 Flux Lora 模型,可用于文本到图像的生成。它借助 Replicate 平台进行训练,为图像创作带来独特风格。
🚀 快速开始
触发词
使用 RTMI
来触发图像生成。
使用 🧨 diffusers 库
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('lichorosario/flux-lora-rtmi', weight_name='lora.safetensors')
image = pipeline('your prompt').images[0]
更多详情
如需了解更多细节,包括 LoRA 的加权、合并和融合等操作,请查看 diffusers 中加载 LoRA 的文档。
📦 安装指南
暂未提供具体安装步骤,可参考上述代码示例进行使用。
💻 使用示例
基础用法
from diffusers import AutoPipelineForText2Image
import torch
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('lichorosario/flux-lora-rtmi', weight_name='lora.safetensors')
image = pipeline('your prompt').images[0]
高级用法
# 可根据文档中提供的示例提示词,替换 'your prompt' 以生成不同风格的图像
# 例如:
prompt = 'RTMI style. Guybrush threepwood, a tall man with blonde hair, wearing a blue pirate coat with gold accents. He has a white shirt underneath, a belt with a gold buckle, and dark pants. His expression is thoughtful, and he has a slight stubble on his face, adding to his adventurous appearance. Guybrush is programming with many computers. Cyberpunk style.'
image = pipeline(prompt).images[0]
📚 详细文档
模型信息
属性 | 详情 |
---|---|
模型类型 | 文本到图像的 LoRA 模型 |
基础模型 | black-forest-labs/FLUX.1-dev |
训练数据 | 《重返猴岛》(Return to Monkey Island)游戏截图 |
推理参数 | 宽度:1024,高度:1024 |
示例提示与输出
示例标题 | 提示词 | 输出图片 |
---|---|---|
Guybrush programmer | RTMI style. Guybrush threepwood, a tall man with blonde hair, wearing a blue pirate coat with gold accents. He has a white shirt underneath, a belt with a gold buckle, and dark pants. His expression is thoughtful, and he has a slight stubble on his face, adding to his adventurous appearance. Guybrush is programming with many computers. Cyberpunk style. | samples/guybrush-programmer.png |
Tv nest | RTMI style. An empty interior scene just inside the entrance of a towering building constructed and inhabited by cockroaches. The perspective is distorted and cartoonish, with the camera positioned as if you're standing just past the grand, slightly arched door that leads into the building. The interior is a labyrinth of narrow, winding corridors that twist and turn unpredictably, with passageways leading up, down, and in all directions, creating a sense of disorientation. The ceiling, walls, and floors are all pristine white, with the floor retroilluminated, casting a soft glow upwards. The hallways have a sleek, futuristic appearance, but are still dimly lit, with shadows that accentuate the strange angles of the walls. The retro-style TVs embedded in the walls have bright orange plastic casings, while the screens emit a soft, white light. The screens are tilted at odd angles, flickering with static or displaying quirky, old-school TV shows, adding a nostalgic glow to the cold environment. The staircases are modern and minimalist, without handrails, their black steps creating a stark contrast against the white surroundings. The stairs spiral upward in erratic directions, some leading to higher floors, while others abruptly end in dead ends. The stairways are crooked, with steps of varying heights and widths, enhancing the building’s surreal, disjointed atmosphere. The white walls are adorned with subtle, recessed cockroach logos, creating an embossed effect that adds texture and detail to the otherwise smooth surfaces. Additional doors are scattered throughout the corridor, some leading to hidden rooms, while others open into completely dark tunnels. The hallways also bifurcate in places, with some passages leading into these pitch-black tunnels, creating a sense of mystery and unease. The entire scene is bathed in a mix of dim lighting from the ceiling, the soft glow from the floor, and the eerie light of the orange TVs, creating a mysterious and slightly unsettling atmosphere. | samples/tv-channel.png |
Restaurant outdoor | RTMI style. textured paper collage The paper texture is very visible. There are different textures. The papers are handmade painted with watercolor and crayons. An empty stylized game background scene using muted colors, mainly blue, yellow, orange, black and violet. A winding path crosses the screen, curving from the top left to the bottom right. Below the path, whimsical trees with irregular, bulbous shapes and playful, swirling bushes fill the scene. To the right stands a cozy, old-fashioned restaurant with large glass windows, through which warm, soft light spills onto the empty street. Inside, simple wooden tables and chairs are arranged neatly, but the restaurant is vacant, giving it a peaceful, quiet vibe. The restaurant's exterior features a large, aged wooden sign, slightly crooked, with an oversized fork carved into it and the text "Crespín Restorán" in quirky, uneven letters. Outside, a few tufts of grass poke through the cracks of the weathered sidewalk tiles, adding to the rustic charm. The night sky above is dark, with a crescent moon shining softly, casting gentle shadows. The street is empty and lonely, illuminated by a few antique-style street lamps that emit a dim, warm light, contrasting with the cold, bluish light of the street itself. The lamps are bent and irregular, with wavy lines, and their light creates playful, wavy reflections on the pavement. The visual style in these images can be characterized as highly exaggerated and car. | samples/restaurant-outdoor.png |
Beach | RTMI game background design. Fisheye lens panorama background for a point-and-click game. The horizon is curvy, and the scene is viewed from a low angle on a deserted beach with soft, golden sand. In the foreground, gentle waves of the deep blue ocean lap against the shore. Far in the distance, a steep cliff rises dramatically from the beach, and perched on top is a very large wooden cabin with a rustic, adventurous design. The cabin features a thatched roof and sturdy log walls in various shades of brown. A winding dirt path carves its way up the cliffside, connecting the beach with the cabin above. Below the cliff, a small wooden pier extends into the water, where a boat is gently swaying in the breeze. On the distant horizon, a small picturesque village resembling Capri is nestled on a hill, its buildings barely visible against the bright, sunny sky. The entire scene is peaceful and tranquil, with the sun casting a warm light over the landscape, while the sense of adventure lingers in the air from the towering cliff and the solitary wooden cabin above. . This visual style is defined by a strong emphasis on exaggerated forms and dynamic compositions. The shapes throughout are distorted and twisted, moving away from realistic representations and instead creating a world that feels fluid and constantly in motion. Buildings and objects often appear bent, curved, or skewed, giving the environment a surreal quality that emphasizes a playful and imaginative tone. The architecture seems to defy gravity, with structures leaning at odd angles, contributing to the overall whimsical feel. Perspective is another key aspect of this style, where traditional rules are intentionally bent or broken. Instead of using a realistic vanishing point, the perspective often resembles a wide-angle or fisheye lens, with objects stretching and warping as they recede into the distance. This creates a sense of depth that is exaggerated and immersive, drawing the viewer into the scene while maintaining a distinctly stylized and cartoonish appearance. Lighting is used to define the shapes and enhance the sense of depth and dimension. Highlights and shadows are applied in a bold, almost graphic manner, bringing out the contours of the objects and creating a strong contrast between different elements. This approach adds to the overall dynamism of the composition, making the scenes feel energetic and full of life. The compositions themselves are carefully structured to guide the viewer’s eye across the scene. The use of exaggerated perspective and distorted shapes creates a visual flow that leads from one area to another, encouraging exploration and engagement with the environment. This sense of movement within the static image helps to reinforce the idea of a world that is in constant flux, where nothing is quite as it seems. Overall, the style focuses on distortion, dynamic perspective, and strong lighting to create an engaging and surreal visual experience. The world feels lively and imaginative, with every element contributing to a cohesive and playful aesthetic that invites the viewer into its unique and fantastical universe. | samples/beach-01.png |
Cliff | RTMI illustration style. View from a cliffside lookout, looking down towards large rocks and the ocean below. The edge of the cliff is covered in rough, gray rocks with small cracks and patches of resilient vegetation growing between them. The deep blue ocean below gently crashes against the rocks at the base of the cliff, creating swirls of white foam. Some of the rocks are partially submerged in water, while others jut out, weathered by the waves. On the right, the ocean extends toward the horizon, with the sunlight reflecting off the water, creating bright glimmers. The sky above is clear, with soft sunlight bathing the scene. The atmosphere is peaceful, with soft shadows cast on the rocks and water. The base of the cliff is composed of dark, rugged rocks with patches of algae that catch the light, while gentle waves leave traces of foam as they break against the rocks. . The visual style in these images can be characterized as highly exaggerated and cartoonish, with a playful use of perspective and distortion. The style features bold and vibrant color schemes, where each area of the scene uses a limited but intense palette, allowing the shapes and forms to stand out sharply. These colors are often applied in large, flat planes, giving the visuals a somewhat graphic novel-like quality, even though they are rendered with a painterly texture. The perspective is notably warped, creating an almost fisheye lens effect in some cases. This warping is especially evident in how walls and floors curve unnaturally, as if the viewer is seeing the scene through a distorted lens. This creates a surreal and exaggerated space that enhances the whimsical and absurd nature of the scenes. Shapes are angular, with a mix of sharp and flowing lines. There’s a clear emphasis on abstraction, where objects and architectural elements are simplified to their most recognizable forms but are bent and twisted to fit the playful tone of the scene. The use of negative space is also significant, where shadows and silhouettes create dynamic contrasts, further adding to the overall exaggerated aesthetic. The perspective does not follow strict rules of realism. Instead, it is dynamically altered to guide the viewer’s eye across the scene, often leading to a sense of depth through exaggerated angles. The lens-like distortion pulls the edges of the scene outward, creating a sense of expansion, as if the environment is stretching away from the viewer. This helps to create a visually engaging and immersive experience, despite the flatness of the color application. Overall, the style combines vibrant, exaggerated color with distorted shapes and a playful approach to perspective, resulting in a visually unique and whimsical world that feels both surreal and immersive. The style is a collage that emphasizes texture and a handmade feel. Use simple geometric shapes cut from textured paper, incorporating a mix of rough, uneven edges to mimic the look of hand-torn paper. Incorporate crayon or pastel-like effects on some of the elements, adding a slightly uneven, grainy texture. The colors should be soft and muted, with subtle variations in hue, as if lightly shaded or colored by hand. The background should have a textured, almost sponge-like appearance, with visible brush strokes or speckled effects. Overall, the piece should have a playful, handcrafted look, combining the rough, tactile qualities of traditional collage with the expressive, childlike feel of crayon or pastel artwork. | samples/punta-01.png |
Singers | RTMI cartoon character design. The image shows two singers passionately performing on stage, both holding microphones and singing with great emotion. The man on the left is bald, with a neatly trimmed beard and wearing a dark blue suit jacket over a white shirt. His head is tilted slightly upward as he sings, with his mouth wide open, conveying intensity and energy. His right hand grips the microphone tightly, and his left hand is extended outwards, as if emphasizing the power of his performance. To his right is a woman with long, curly blonde hair cascading over her shoulders. She is wearing a black, sparkling dress with a deep neckline, and she too is singing with enthusiasm. Her expression is powerful, with her mouth open wide as she belts out a note. She holds her microphone in her right hand, while her left hand is extended in a dramatic gesture, as if reaching out to the audience. She wears a watch on her left wrist and a ring on her right hand, adding to her glamorous appearance. The background features a stage with a decorative backdrop that includes a large circular design in shades of yellow and orange, resembling an abstract sun. Overhead, green stage lights illuminate the scene, casting a vibrant glow on the performers. Behind them, a dark curtain provides contrast, making the singers stand out even more. The atmosphere is lively, with the lighting and their dynamic expressions capturing the essence of a passionate performance. | samples/singers-01.png |
Spiderman | RTMI style. illustration of spiderman at the comic con | samples/spiderman.png |
Monna Lisa | RTMI style. character design. The image depicts Monna Lisa by leonardo da vinci in cartoon RTMI style. | samples/monna-lisa-03.png |
室内男子 | RTMI cartoon style character design. This image captures a casual indoor scene, likely taken in a clothing store or a wardrobe area filled with various garments. The focal point is a caucasian man standing in the center, smiling broadly with a friendly, relaxed expression. The overall style evokes a playful, almost retro vibe. The man is wearing a tight-fitting, long-sleeved gray shirt, which closely follows the contours of his body. The man has sweaty armpits in darker color. Around his neck, he is wearing a tan and beige checkered scarf, loosely wrapped but still snug enough to stand out as a fashion accessory. The scarf's texture appears soft, perhaps woolen or knit, and it adds an element of warmth and contrast to his otherwise sleek outfit. In the background, the space is organized with shelving and clothing displays. On the left, a shelving unit holds a few folded garments, including shirts or sweaters, neatly stacked in different colors. Next to him on the countertop is a black shopping bag adorned with white and green designs and text, possibly from the store itself. A vibrant green ribbon is attached to the bag, adding a pop of color. Behind him, part of the shelving has decorative items, including what looks like a small piece of art or signage featuring bold black and white graphics. To the right of the man, a bright green curtain is partially visible, which may be a fitting room or divider within the space. The lighting in the room is moderate, with a soft glow illuminating the scene evenly, giving it a clear, well-lit appearance without harsh shadows. The atmosphere feels casual and perhaps a bit playful, given the man's fashionable and somewhat whimsical look. the man's hairstyle is a striking feature, characterized by a mix of textures that gives it a carefully styled yet relaxed appearance. His hair is dark, likely black or very dark brown, with a healthy, shiny finish. The style is divided into two distinct textures: a sleek, straight fringe (bangs) at the front, and voluminous, professionally styled curls throughout the rest of the hair. The fringe is straight and falls neatly over his forehead, with several strands parted slightly in the middle. The bangs are relatively long, reaching just above his eyebrows. The hair is smooth and flat against his forehead, providing a sharp contrast to the more voluminous curls behind it. The straightness of the fringe is likely intentional, adding a polished, elegant touch to his look. The rest of his hair, in contrast, is full of soft, bouncy curls, similar to those created with a curling iron or from a salon blowout. These curls start at the sides of his head and flow outward, adding significant volume and a lively texture. The curls are medium in size, not too tight but defined enough to create a dynamic, flowing look. The back of his hair maintains this voluminous curl pattern, rounding out the shape of his hairstyle and giving it a balanced, full appearance. This blend of sleek, straight bangs with voluminous curls creates a unique juxtaposition, adding both sophistication and a touch of playfulness to his overall style. The hairstyle feels modern but with a nod to vintage influences, combining the precision of straight bangs with the soft glamour of big, flowing curls. It’s clear that this look has been styled with care, likely in a salon setting, to achieve this distinct mix of textures. The curls go down the shoulders | images/example_b7k8vrb7y.png |
📄 许可证
本模型使用 flux-1-dev-non-commercial-license 许可证。
Stable Diffusion V1 5
Openrail
稳定扩散是一种潜在的文本到图像扩散模型,能够根据任何文本输入生成逼真的图像。
图像生成
S
stable-diffusion-v1-5
3.7M
518
Stable Diffusion Inpainting
Openrail
基于稳定扩散的文本到图像生成模型,具备图像修复能力
图像生成
S
stable-diffusion-v1-5
3.3M
56
Stable Diffusion Xl Base 1.0
SDXL 1.0是基于扩散的文本生成图像模型,采用专家集成的潜在扩散流程,支持高分辨率图像生成
图像生成
S
stabilityai
2.4M
6,545
Stable Diffusion V1 4
Openrail
稳定扩散是一种潜在文本到图像扩散模型,能够根据任意文本输入生成逼真图像。
图像生成
S
CompVis
1.7M
6,778
Stable Diffusion Xl Refiner 1.0
SD-XL 1.0优化器模型是Stability AI开发的图像生成模型,专为提升SDXL基础模型生成的图像质量而设计,特别擅长最终去噪步骤处理。
图像生成
S
stabilityai
1.1M
1,882
Stable Diffusion 2 1
基于扩散的文本生成图像模型,支持通过文本提示生成和修改图像
图像生成
S
stabilityai
948.75k
3,966
Stable Diffusion Xl 1.0 Inpainting 0.1
基于Stable Diffusion XL的潜在文本到图像扩散模型,具备通过遮罩进行图像修复的功能
图像生成
S
diffusers
673.14k
334
Stable Diffusion 2 Base
基于扩散的文生图模型,可根据文本提示生成高质量图像
图像生成
S
stabilityai
613.60k
349
Playground V2.5 1024px Aesthetic
其他
开源文生图模型,能生成1024x1024分辨率及多种纵横比的美学图像,在美学质量上处于开源领域领先地位。
图像生成
P
playgroundai
554.94k
723
Sd Turbo
SD-Turbo是一款高速文本生成图像模型,仅需单次网络推理即可根据文本提示生成逼真图像。该模型作为研究原型发布,旨在探索小型蒸馏文本生成图像模型。
图像生成
S
stabilityai
502.82k
380
精选推荐AI模型
Llama 3 Typhoon V1.5x 8b Instruct
专为泰语设计的80亿参数指令模型,性能媲美GPT-3.5-turbo,优化了应用场景、检索增强生成、受限生成和推理任务
大型语言模型
Transformers 支持多种语言

L
scb10x
3,269
16
Cadet Tiny
Openrail
Cadet-Tiny是一个基于SODA数据集训练的超小型对话模型,专为边缘设备推理设计,体积仅为Cosmo-3B模型的2%左右。
对话系统
Transformers 英语

C
ToddGoldfarb
2,691
6
Roberta Base Chinese Extractive Qa
基于RoBERTa架构的中文抽取式问答模型,适用于从给定文本中提取答案的任务。
问答系统 中文
R
uer
2,694
98