blue archive2 krea2 — Mô hình AI KREA2

KREA 2 V2 Update: Dataset overhaul, trained 10 epoch each on 512/768/1024. Dim set to 64 for test. The halo should be easy to remove by reduce LoRA weight. The best balance is 0.8 in my test. Trigger word has changed like official Krea 2 loras: azure youth anime art style But you should be able to use it without trigger word, cuz it's pretty weak. It can be used as a high-contrast or flat color style module. Use proper prompt like . Still, the model do not contain any character information. You still need character lora if you want a typical character. ==================== KREA 2 V1 Update: I've back for a little test, trained on my dataset with NL captions. Trained on my own 4090. Weight 1 is fine, Trigger word is BArtstyle style USE WITH NATURAL LANGUAGE. DANBOORU TAG IS NOT WORKING. If it doesn't work well, try remove words like "anime, cartoon" in the prompt. It may slightly affect the style. Halo may undesirably come out in certain prompts. If happens, try a different prompt or reduce lora weight, or just try another seed will help. This model do not contain any character information. You still need character lora to make it work properly, this is limited by Krea 2's base model. I will test it after I train my first character lora for Krea2 and update info here. Info below has expired. Time flies huh? It has been 2 years since I train model. =============================================== I've wasted too much time on this model... Probably this will be the final update, I may focus on SDXL when 4090 be affordable to me, or just fixing other LoRA I made if 4090 is still cost way too much in next half year. The original decision was using dreambooth for update, but I'm using 3060 so the time effort is not good. So, still LoRA. REv3 Change Log： - Base model now have 3 versions, which have different feature: - CF3 - Have a good balance between replica art style and user-friendly. - You should use it on CounterFeitV3 (and those CF3's mixed model, e.g. PrimeMix/MixPro) - DO NOT USE IT WITH Anything SERIES. I tried and it looks awful. - Silicon29 - Trained on Silicon29-Dark, it sacrificed a bit BA's clean line art, but it has a more BA-ish vivid color+god ray than CF3 version. - I strongly recommend this version (at least you should try it out!) - It has a better variety with non-base model when still kept the base model's characteristic. - Pink theme is its weakness. If you want to generate the BA Finale's blue-pink gradient scenery, use CF3/ANY version instead. - - ANY - Trained on ANYv5, harder to use than CF3 or Silicon29, but it seems to be the most BA one. Only suggest when you using AnythingV5 or those model based on it. - Dataset rehauled - Manually downscale + denoise before send it to bucket to avoid aliasing. - Trained resolution is 1024x768. So it's better generate horizonal image than vertical one. Suggest generate resolution : - 432x768 (Vertical, x2.5 for 1080P) - 960x544 (Horizonal, x2 for 1080P) - Concept change - now the typical BA style (blue sky+ modern city / ruined classroom + water) has its own concept. - trigger word: blue sky - blue theme now can generate nighttime. - Preivew (test on Silicon29 ver.) - - Block Weight Training - Use a more conservative set. So these LoRAs will affect more anatomy but not sacrificed too much effect. - While using REv3, your LoRA weight can up to 1, since the anatomy is basically hopeless XD - joke aside, v9b has sacrificed too much color for meh anatomy while STILL CAN NOT use at a higher weight. - REv3's anatomy is better than v7 I guess. - I've tested some sampler. - For conclusion: Restart ≈ DPM++ 2M Karras > DPM++ 2S a Karras > any sampler with SDE - UniPC is basically a downgrade version of Restart. - DPM++ 2S a Karras is oversaturated on typical model and weight. not bad but just not BA style, you should use a lower weight on this sampler. - Restart & DPM++ 2M Karras will cause blue or green spots or something like jpeg artifact in typical situation. It can not be fixed cuz it's base model's problem - Although CounterFeit V3 provide a relatively good art, when still focus on prompts, the model itself has been overtrained (I believe that's why it's very picky on VAEs.) - ↑ that's why you should use CFv2.5's VAE while using this lora models. ================== V9b Change Log： - Base model changed to CounterFeit V3. - Use Block Weight Training. Anatomy should be way better than V7. - BAPV no longer used as trigger word. You can still use color scheme to trigger typical color. - For the best result, you should choose DPM++ 2S a Karras sampler & use 4x-ultrasharp as upscaler. Known Issue： - Color shifting has been noticeably sacrificed for better anatomy. Check the last preivew for more info. V9b will affect less on the whole image's composition and charactor posing. I might release a version without Block Weight Training, which has better colors, but you should know how to properly use LoRA Block Weight. - It tends to give a high-contrast two-tone shadow even on nighttime. Consider combine with if you need to generate dark scene. - Bad Composition still occurs if you are using a relatively high weight (over 0.7), consider use LoRA Block Weight (LBW) with OUTALL setting to reduce the affect. If you use LBW, higher weight up to 1 is Okay. V7: Since the training data is based on Blue Archive's PV & CG screenshot, this LoRA do not very capable with NSFW content. V9 might be better. But I haven't tried it out. Feel free to test it by yourself! Weight 0.6-0.7 is good enough. About 0.4-0.8 is working zone. Higher will have a better color scheme but worse anatomy. ================== Support color scheme in grid: (V7 sample) Still, it only works perfectly without any charactor LoRA. You can get similar result with charactor LoRA, but you need to be patient. V9b has been MUCH better, but still you need to roll dices. The method below is only for older LoHA version. ====

Từ kích hoạt

azure youth anime art style

3,102 lượt tải · 212 lượt thích trên Civitai

SHA256

9C19DC382D92002BFC53ECE7DF5556ACC288579AF9F0719752059DFAE7FE9F70

ID mô hình ModelTrends.ai

#26956

Báo cáo nội dung

Ảnh mẫu

Prompt và thiết lập do những người tạo ra các ảnh này trên Civitai chia sẻ. Chọn một ảnh để xem nó được tạo như thế nào.

Lấy prompt này

Ảnh của mmdzz trên Civitai

Báo cáo

Thiết lập

Sampler

Euler beta

Số bước

Guidance

Seed

318665869269900

Kích thước

2048x3072

Clip skip

Trọng số đã dùng

Prompt

azure youth anime art style, @M4LPH4SX, , A young woman with short, choppy white hair and striking teal eyes stands against a wall covered in large, jagged black graffiti letters. She has a bold, playful expression, winking one eye and sticking her tongue out while raising her hand in a peace sign. She wears a black cross-shaped hair clip, multiple ear piercings, a black choker, and layered gold necklaces. Her outfit is a mix of edgy streetwear: an oversized, bright cyan sweatshirt with a black skull and dragon graphic, worn off one shoulder to reveal a red strap. She wears a short, pleated purple plaid skirt with a black garter strap on her thigh, and distressed black thigh-high stockings with large rips. Her chunky black boots have red laces and accents. A small red and black bag hangs at her side. The color palette is bold and high-contrast, dominated by the vivid cyan of her sweatshirt, the stark black of the graffiti and her accessories, and the pale tones of her skin and hair, with pops of red and purple. The artwork is rendered in a crisp, graphic anime style with flat shading. The overall mood is one of rebellious, energetic youth, a confident and stylish figure in an urban, grunge-inspired setting., <lora:BArtstyle-KREA2-V2-1024:1.0> <lora:MALPHAS-KREA-V1 (1):1.0>

Ảnh của NEIPA trên Civitai

Báo cáo

Thiết lập

Sampler

Euler

Số bước

Guidance

Seed

915101735339493

Kích thước

1288x1936

Prompt

azure youth anime art style. Four young women in sailor uniforms pose for a selfie, captured from an overhead perspective. The image is taken at eye level with the subjects huddled closely together and looking toward the camera. SUBJECT A (top left): A woman with short black hair and round glasses gives a thumbs-up gesture with her right hand while wearing a white sailor uniform with navy stripes on the sleeves and collar, part of which appears to be a school uniform. She is in the upper-left corner, slightly above center, with her head tilted down toward the camera. SUBJECT B (top): A woman with dark hair pulled back by a red band makes a playful face—winking one eye and sticking out her tongue—with her left hand raised to make a peace sign or V-sign gesture. She is positioned at the top of the frame, centered slightly above Subject C’s head. Her uniform has navy stripes on the sleeves and appears to be part of the same school uniform. SUBJECT C (center): A young woman with short black hair and bangs smiles broadly while resting her right hand against her cheek. She wears a white sailor uniform with matching navy-striped cuffs. Her head is tilted toward the camera, facing directly forward. SUBJECT D (right): A woman with dark hair and a wide smile covers part of her face with her right hand in a playful gesture while wearing a white sailor uniform with navy-striped sleeves. She is positioned at the top-right, smiling directly toward the camera. The selfie is taken from above using a downward-facing camera angle likely held by Subject C’s arm extending into the frame from the bottom center (the forearm and hand are partially visible). The group forms a tight cluster in the foreground with their heads close together. In front of them: A typical school desk is positioned on the left side. Behind them: In the background, there’s a table with various items including papers, snacks (a beige snack bag), and a green chair to its right. Further back are more chairs, some green, in what appears to be a backstage or cafeteria area filled with equipment and personal belongings like a black backpack or case on the floor. All four subjects wear similar white sailor-style uniforms with navy blue stripes at the cuffs and collar, suggesting they are part of a school drama club or theater group. The uniform includes elements such as sailor collars and striped sleeves. The lighting appears to be from above, casting soft shadows on their faces and creating a slightly warm indoor atmosphere with no harsh contrasts. All subjects are smiling or making playful gestures for the photo. The selfie is framed such that it captures them closely together in a tight cluster, indicating they are posing as a group for fun or promotional purposes.

Ảnh của Choco_Tarte trên Civitai

Báo cáo

Thiết lập

Sampler

Euler a

Số bước

Guidance

Seed

296212169279876

Kích thước

3240x3456

Prompt

[masterpiece, newest,absurdres,incredibly absurdres, best quality, amazing quality, very aesthetic, 1 girl, white cat ear, blue eyes,thin white cat tail,hair between eyes, medium hair, wavy , white hair, messy hair, medium breasts,fair skin,Red plum flower hair ornament,,solo,@kastusika hokusai, shunga , flowing blush stroke turning into ukiyo-e, sitting on red fabric bench subject: portrait, from side, oiran, clevage, black kimono with red edging, holding kiseru , downcast eyes , looking away, The contrast between the realistic depiction of the figures and the painterly backgrounds and foregrounds creates a mysterious effect. Smoke rises from the pipe.furisode, no cheongsam, hanbok, sari, draped cloth, lederhosen, no sash, no hair accessory, silkm0nm0n., flat colors with thin, elaborate linework., moroboshi kirari<lora:Mrpopo anime full potential:1.0> <lora:monmon mon2133 style KREA V1:1.0> <lora:Moroboshi_Kirari_KREA2_V1:0.8> <lora:BArtstyle-KREA2-V2-1024:0.2>

Prompt phủ định

worst quality, low quality, normal quality, score_1, score_2, score_3, artist name, bad anatomy, bad hands, extra fingers, missing fingers, fused fingers, extra toes, missing toes, extra limbs, malformed limbs, blurry, lowres, jpeg artifacts, uncanny, lowres, bad anatomy, worst quality, low quality, normal quality, bad hands, mutated, extra fingers, artifacts, disfigured,

Ảnh của NEIPA trên Civitai

Báo cáo

Thiết lập

Sampler

Euler

Số bước

Guidance

Seed

875013617708309

Kích thước

968x1344

Prompt

masterpiece, best quality, very aesthetic. @NJSW33T. azure youth anime art style. ArsMJStyle, Colored pencil hyperdetailed realism. SUBJECT A (center): A beautiful Japanese high school girl with short, dark hair and thick-framed white sunglasses, wearing a yellow vinyl raincoat over her Japanese school uniform. She stands upright facing the camera directly with a neutral expression, head tilted slightly forward but not lowered or tilted to the side. Her right arm is fully extended straight toward the camera in a finger-gun gesture: index finger pointed directly at the lens and thumb raised vertically upward while all other fingers are curled into her palm; her hand is prominently positioned in the foreground. She wears thick white-framed sunglasses that sit naturally on her face, covering her eyes completely with visible, clear white frames. Her school uniform appears to be a standard Japanese high school outfit under the yellow vinyl raincoat, which has a smooth, slightly glossy appearance characteristic of thin yellow vinyl material. Her fingernail color is a normal flesh tone. SUBJECT B (center): A ginger tabby cat with soft and detailed fur, being held in the girl’s left arm; it looks directly toward the camera with a warm, exuberant, open-mouthed smile. Its eyes are half-closed in a cheerful eye-smile, with the upper eyelids gently lowered and the eyes still visibly open, creating a lively, delighted expression rather than a sleepy or drowsy look. Its lips are curved upward at the corners in a clear happy grin, with the teeth slightly visible. The cat is nestled against her body, its front paws resting on her coat or shoulder area while its back legs are wrapped around her torso for support. All four of the cat's legs are visible. Camera geometry: frontal horizontal viewing angle at approximately 0 degrees, camera height slightly above subject level but not upward-looking; no Dutch angle (roll = 0°), optical axis pointing directly toward the subjects’ faces. Perspective is close-up with a shallow depth of field, framing the girl and cat centered in the shot. Environment: The background features a weathered gray-blue wall or doorframe to her right, possibly with a faded sign or utility box visible; a dark greenish doorway or entrance area on the left side. The setting appears urban or residential, likely outdoors near a building entrance. Lighting: Natural daylight illumination casting clear shadows and highlights; bright region centered on the girl’s face and cat, suggesting direct sunlight from front-facing angle with no strong shadow direction apparent beyond minor contrast.

Ảnh của Enjukun trên Civitai

Báo cáo

Thiết lập

Sampler

euler

Số bước

Guidance

Seed

886785949775972

Kích thước

1664x2496

Prompt

2.5D anime illustration style, smooth gradients, thin faint lineart, saturated at the core with a soft bloom, brightest underneath and behind, her eyes self-illuminated, glossy with bright catchlights, shadows that keep their color, an equal part of the artworkRe_Zero, Rem, short blue hair covering one eye, bright blue eyes, pale skin, slender build, a single small pale horn on her forehead, black and white maid dress with a frilled apron and a ribbon headband, Heavy black smoky eyeshadow, dense and opaque, covers the entire eyelid up to the brow bone, with thick dramatic black eyeliner ending in a long sharp point past the outer corner and thick black mascara on dramatic false lashes. The lips are full and rounded, painted solid black in a thick opaque coat, the lacquered shine catching light as sharp bright highlights. Bold high-impact makeup, heavily applied, a black LED display placard with red letters held over her shoulder reading "BUZZ FOR ENJU'S TOY"Buoyant Signal; Rem, black and white maid dress with frilled apron and ribbon headband, one hand lifting cigarette to lips, other hand holding rectangular LED board forward, wrist bent, board reads "BUZZ TO MARK HER, " eyes half-lidded, slow deliberate gaze, jaw lifted, faint smirk, smoke pulling outward, shoulders relaxed, spine arched, waist twisted, weight on one leg, ribbon tilted, frilled apron catching motion; a towering upside-down chandelier forest rooted in cloud, each crystal cluster hanging from the sky and dripping upward into the sea below, the LED board glowing its message across the grove, the cigarette ember pulsing small orange light; cold white from the board cuts through translucent pink mist, while warm amber smoke drifts through and catches icy blue chandelier glints; sketch-line rendering with soft airbrush gradient; deep black and stark white for the dress and forest geometry, saturated red for the LED text and ember, muted rose for the mist; finest detail on the ribbon headband, frilled apron texture, and smoke curl, softest in the chandelier haze and cloud edges; centered full body, hips angled sideways, world frames her in converging crystal lines that lead the eye directly to her face and the lit board

Prompt phủ định

bad anatomy, bad hands, watermark, ugly, distorted, censored, lowres, pixelated, distorted, jpeg artifacts, signature, (watermark), overexposed, underexposed, missing fingers, extra fingers, missing limbs, extra limbs, bad eyes, cloned face

Mô hình tương tự

instagram influencers

Mô hình AI để tạo hình ảnh, video & văn bản

blue archive2 krea2 là một LoRA thuộc họ Krea2 được liệt kê trên ModelTrends.ai, danh mục chỉ đọc về các mô hình AI hình ảnh, video và văn bản mã nguồn mở. So sánh nó với các mô hình Krea2 khác, xem điểm độ hot để biết xu hướng và mở trên PirateDiffusion hoặc BitVector để dùng thử.

Thêm mô hình Krea2 Thêm LoRA Krea2 anime gaming artstyles style game art artstyle blue archive

Duyệt theo họ mô hình