📢Tsubaki.3 — Tell Us What You Think! — 📝 【Tsubaki.3 — We Want to Hear From You! Survey Now Open】 Thank you for always using PixAI! It's been one month since th…
View details
Comm.DiTのテスト１５
１４の続きです。 １３に追加して、説明書に記載のあった ※説明書とは、モデルおよびLoRA両方の説明書を読んでいます。主にLoRAの説明を反映させています。 １・ＬｏＲＡの強さを０．９まで絵引き上げ。（前は０．７） ２・プロンプトに「anime coloring,」を追加。 １３にくらべて、色彩とキャラがくっきりと表現されてきました。 設定を細かくいじるここで、ここまで変化するのか。と思いました。 Comm.DiTシリーズはユーザー様が調整したモデルとのことで、 調整したユーザー様の個性が色濃く出るのが面白いところです。 PixAIの運営さんが調整したTsubakiモデルとは、また別の遊び方だと 感じました。 PixAIさんのモデルは、細かい設定を省いてわかりやすく生成できる感じですが、 Comm.DiTは、細かくステータスがいじれる一方で、その分難度 が上がるモデルということなのかもしれません。 ただ、Sampling Stepsは低くても、きれいなイラストが仕上がるような作りもできるようなので、そこがモデルを作成される、ユーザー様の腕の見せ所なのかもしれません。 今回遊ばせていただいたモデルは、Sampling Steps１２で、４枚生成で４２００Ｋでした。コストが安価で、好きな感じのアニメイラストが作れました。 しっかり説明書があるモデルやＬｏＲＡだと分かりやすくていいなと思いました。 遊んでいて面白いです♪ Comm.DiT Test 15 Continuing from Test 14... Building on Test 13, I applied the following changes based on the documentation (I consulted the instructions for both the model and the LoRA, though I primarily incorporated the LoRA's guidelines): 1. Raised the LoRA strength to 0.9 (previously 0.7). 2. Added "anime coloring," to the prompt. Compared to Test 13, the colors and the character are rendered much more sharply. It really struck me how much of a difference fine-tuning the settings can make. Since the Comm.DiT series consists of models adjusted by users, it’s fascinating to see the creator's unique style shine through so clearly. It feels like a different kind of experience compared to the "Tsubaki" model adjusted by the PixAI team. PixAI’s own models feel designed for easy generation without needing complex settings, whereas Comm.DiT allows for granular control over parameters— which perhaps makes it a model with a steeper learning curve. However, it seems possible to produce beautiful illustrations even with low Sampling Steps, so that might be where the creator's skill really comes into play. For this test, I generated four images using 12 Sampling Steps at a cost of 4,200K. It was affordable, and I was able to create anime-style illustrations that really appealed to me. I appreciate models and LoRAs that come with clear documentation; it makes the process much easier to understand. It’s really fun to play around with! ♪
Show Parameters
- Original Prompt
- Character A and Character B are sitting across from each other in a café. They are chatting happily while looking at each other's faces. There is a strawberry shortcake in front of Character A. There is a cheesecake in front of Character B. Character A:,kagamine rin, solo, blonde hair, short hair, blue eyes, Petite, hair ornament, hairclip, detached sleeves, black shorts, sailor collar, skindentation, bow, shirt, leg warmers, Character B:,Megurine Luka Sour uniform,hair ornament,hairclip,brooch,black dress,sleeveless dress,wrist cuffs,black thighhighs,asymmetrical footwear,thigh boots,black footwear, anime_screenshot,anime coloring,
- Size
- 1536 x 1536
- Sampling Steps
- 12
- Sampling Method
- Euler
- CFG Scale
- 1.1
- Negative
- worst quality, photorealistic, bad anatomy, blur, low resolution
Model & LoRA used
Similar Images
Show more
ボカロのみんなでカフェタイム♪
祐二
Test
Selodiomivus.
c
Luka & Miku
chokunnaluck
d
character: kagamine_len, kagamine_rin, copyright: vocaloid
1boy, 1girl, 10-year-old twins, siblings, childlike proportions, small stature, cute round faces, short boy, short girl, blonde hair, spiky hair, tiny ponytail, short bob cut, large white hair bow, hair ornament, blue eyes, white headphones, headphones on head, collared shirt, white shirt, yellow necktie, black shorts, grey shorts, indoor, small wooden table, sitting at table, teacup, saucer, plate of cookies, cookie, food theft, reaching, leaning forward, mischievous grin, surprised expression, puffed cheeks, medium shot, slightly from above, three-quarter view, 2d anime style, cel shading, clean lineart, soft colors, high detail
Seen from a slightly elevated three-quarter view, two ten-year-old twin siblings with childlike proportions, petite stature, and soft, rounded cheeks sit opposite one another at a small circular wooden table. On the left, the young girl leans enthusiastically forward across the tabletop; she has a chin-length blonde bob crowned with a large, crisp white ribbon bow and wears sleek white overhead headphones covering her ears. She is dressed in a short-sleeved white button-up collared shirt tucked into grey shorts, accented by a bright yellow necktie. With a teasing lopsided grin and one bright blue eye winking, her right fingers playfully pinch and lift a golden-brown round cookie straight from her brother's plate, while her left hand presses flat onto the wood for balance, shifting her weight forward. On the right, the young boy leans back with comical surprise; he features short spiky blonde hair with a tiny tuft ponytail gathered at the nape of his neck, matching white overhead headphones, bright blue eyes opened wide in exaggerated shock, and cheeks puffed out in mock indignance. He wears an identical short-sleeved white button-up collared shirt with a bright yellow necktie, paired with black shorts. His right arm reaches mid-air in a desperate attempt to rescue the snack, while his left hand clutches the edge of the table near his cup. Between them rests a white ceramic plate holding the remaining cookies, flanked by two pastel tea cups on saucers holding warm amber tea. A few tiny loose crumbs tumble straight down from the lifted cookie toward the plate surface, and thin curling ribbons of steam rise from the hot tea. Crisp, clean cel-shading outlines the figures with tidy linework, illuminated by soft, warm daylight that casts sharp, neat shadow shapes across the wooden grain, beneath their sleeves, and under the saucers. The palette balances sunny yellow accents, biscuit-toned golds, and sky-blue irises against soft whites and clean neutral shorts. Playful and heartwarming.
djxkjch
Comments2