谷歌Nano Banana Pro:新手AI影像編輯神器!

Ai




我試用Google全新Nano Banana Pro:夢想中的AI Photoshop來了!

Google為了挑戰OpenAI,一直不斷嘗試各種方法,想吸引用戶由ChatGPT轉用自家Gemini AI。其中一個重要武器,就是他們的AI圖像生成及編輯模型——Nano Banana。自從推出後,Nano Banana迅速爆紅,Google亦在多個平台大力推廣。為了保持熱度,Google最近推出了大幅升級版Nano Banana Pro,我有幸在正式發布前搶先試用,感覺它將會是像我這類圖像編輯初學者的Photoshop替代品。

自8月底Nano Banana首次推出後,我已經用它做過不少事情:從修改現有素材加入文字,到為我們的Authority Insights Podcast製作吸睛縮圖,Nano Banana的確為我的工作帶來極大便利。它在創造和修改圖像方面的表現非常出色,否則也不會爆紅。

不過,沒親身試過,你可能不會知道Nano Banana有時用起來超級令人挫敗。它經常漏掉重要細節,甚至完全誤解你的指令,令你得重頭開始。當你想讓它在後續修改時,往往會徒勞無功,因為它會一再重複同一個錯誤,好像不聽指令一樣。如果你想要非正方形的圖片?祝你好運!

而Nano Banana Pro幾乎解決了以上所有問題。它能更準確理解初始提示,亦能真正接受後續修改要求,最重要的是,它終於支援16:9及其他比例的圖片生成。這是因為它基於全新的Gemini 3 Pro模型,具備推理能力;相比之下,原版Nano Banana是用Gemini 2.5 Flash(官方名稱是Gemini 2.5 Flash Image)。利用具推理能力的模型,令Nano Banana從一個有趣但偶爾令人抓狂的工具,變成真正的AI圖像助手。

初學者也能輕鬆製作迷因(memes)

有些人可能會質疑:AI能做迷因嗎?迷因本質上是人類創作的,不過它們的核心在於廣泛分享和不斷被改編。雖然現有迷因生成模板方便簡單編輯,但有些迷因需要較高的Photoshop技術才能調整圖片格式,很多人(包括我)都不懂這些技巧,只能在旁觀望,羨慕別人花大量時間製作的搞笑圖片。

有了Nano Banana Pro,我幾秒內就能完成這類迷因創作。以下兩個例子展示了Android機械人吉祥物被改編成日本動漫《咒術迴戰》中的「Nah, I’d win」和「Domain Expansion」場景,效果相當生動。

我難以想像手動製作這些要花多少時間,但Nano Banana Pro只用了幾秒。更令人驚喜的是,我還能與Gemini來回溝通,不斷微調第一張圖片,這是原版Nano Banana完全做不到的,否則我得開無數個新對話,不斷修改提示才能達到同樣效果。

終於不再毀臉的圖片編輯

原版Nano Banana最大問題之一,是編輯圖片時常常會扭曲人臉。我曾經讓它移除一張朋友合照的背景物件,結果除了背景乾淨了,所有人的臉竟然變了樣,看起來像完全不同的人!

我擔心Nano Banana Pro會有同樣問題,得小心翼翼下指令。不過很幸運,Nano Banana Pro在編輯人像時不再毀臉,表現非常出色。

例如,我用它把照片中我戴的Galaxy XR頭戴設備換成Apple Vision Pro,甚至重建了部分鼻子,並巧妙地把朋友公寓的反光投射到護目鏡上。當然細看反光不完全吻合實際場景,但沒對比原圖大多數人都不會察覺。

另一張是我手持「Gemini」字樣的薑餅餅乾,指令它把背景人員移除(只保留兩個廚師),並把字改為「Nano Banana」,效果相當逼真,文字看起來像真的用糖霜寫成。

Nano Banana Pro不只是Photoshop替代品

雖然我之前把它比作Photoshop,但我認為Nano Banana Pro的能力更勝於此。Photoshop可以利用AI做加減內容或生成圖片,但Nano Banana Pro基於Gemini 3 Pro,具備真正的推理能力,能分析複雜數據再生成圖像。

舉例來說,我給它我目前的健身計劃,要求畫出鍛煉肌肉的圖示;又給它Android測試版及發布日期,請它畫出時間線,它甚至寫了Python腳本來繪製數據。雖然這些圖還不是百分百完美,仍需微調,但Nano Banana Pro能解讀並視覺化生數據,令人期待未來更多應用。

多種圖片比例支援,使用更靈活

原版Nano Banana頑固地只生成1:1正方形圖片,想要其他比例得用技巧欺騙它。Nano Banana Pro則原生支援16:9、2:1等比例,讓生成的圖片可廣泛用於不同場景,靈活度大增。

AI仍有不足:Nano Banana Pro的挑戰

儘管提升顯著,但Nano Banana Pro仍有不足。生成圖片解析度低於1080p,文字渲染雖然改善,但仍會出錯,尤其是背景文字或提示中未明確指定的字詞。

它也常常無法正確顯示指針時間,這是AI圖像模型的老問題。此外,要產生完美圖像仍需技巧,提示不佳可能會產生恐怖結果。安全機制也限制了涉及公眾人物的修改,像是不能把知名人士臉放到別人身上,這種創意仍需Photoshop等傳統工具。

總結

整體來說,Nano Banana Pro令我印象深刻。它的推理能力讓使用體驗大幅提升,在幾乎所有方面都比原版更好,也不再令人挫敗。對我這種初學者來說,它已經能取代Photoshop,未來我也會更多使用它,不單止是工作。

你又怎麼看Nano Banana Pro呢?歡迎留言分享你的想法!

我的評論與啟發

Google這次的Nano Banana Pro發展,代表AI圖像生成正在從「玩具」向真正生產力工具轉變。以往AI繪圖模型多數只能憑字面指令生成靜態畫面,常出錯且缺乏靈活度,令用戶體驗不佳。Nano Banana Pro透過搭載具推理能力的Gemini 3 Pro,不只理解更複雜的指令,還能進行多輪交互,這是向「智能助手」邁出的重要一步。

這不僅降低了使用門檻,讓非專業用戶也能輕鬆製作專業級圖片和迷因,還打開了AI在教育、設計、數據視覺化等多個領域的應用潛力。尤其是它能將純文字數據轉成可視化圖形,對於需要快速製作圖表的用戶非常有幫助。

然而,AI的限制仍然明顯:解析度不高、文字生成錯誤、時間顯示錯亂等問題提醒我們,AI還未成熟到完全取代人類的審美與判斷。加上道德和法律層面的考量,像是禁止涉及公眾人物的圖像生成,顯示AI技術發展必須兼顧責任與規範。

未來,AI圖像生成工具如果能結合更高解析度、更細膩的文字處理能力,以及更加靈活的互動界面,將有機會真正改變創意產業的生產流程。Google的Nano Banana Pro已經踏出了一大步,其他競爭者也必須加快腳步,否則用戶可能會迅速被這種更智能、更易用的AI工具吸引走。

對香港用戶而言,這類工具不僅能節省時間和成本,更能激發創意,尤其是對小型創作者和中小企業來說,是一個值得關注和嘗試的新選擇。只要能多點中文本地化和調整,Nano Banana Pro未來有望成為香港數碼創作的新利器。

以上文章由特價GPT API KEY所翻譯及撰寫。而圖片則由FLUX根據內容自動生成。

🎨 Nano Banana Pro 圖像生成器|打幾句說話就出圖

想畫人像、產品圖、插畫?SSFuture 圖像生成器支援 Flux Gemini Nano Banana Pro 改圖 / 合成, 打廣東話都得,仲可以沿用上一張圖繼續微調。

🆓 Flux 模型即玩,不用登入
🤖 登入後解鎖 Gemini 改圖
📷 支援上載參考圖再生成
⚡ 每天免費額度任你玩
✨ 即刻玩 AI 畫圖
An ultra realistic portrait of a young man (facial detail 100% matches with the reference photo) relaxing on a wooden balcony with a scenic view of green mountains in the background. Golden-hour sunlight from the left side creates a warm glow, soft highlights on his hair and shoulders, and a subtle sun flare. He is wearing a black t-shirt in a white shirt with long sleeve, a black ripped jeans and black white sport shoes. his short straight hair is blowing by the morning wind. He sits with his body slightly angled, shoulders turned to the right side of the frame, one leg casually folded up onto the chair.

He holds a white cup emitting light steam, suggesting a freshly made hot drink. His face turns gently to the right with a subtle, relaxed smile, giving a calm and peaceful expression. The environment is an outdoor highland area with silhouettes of hills and trees in the distance, softly darkened by the backlit sunlight.

Camera angle is low angle level but slightly from the left side, creating a natural and intimate framing. Background is softly blurred with shallow depth of field, using a wide-aperture lens. Warm, earthy, soft color tones {
"intro": "Create an ultra realistic 8K UHD DSLR photo based on the attached image as a reference of facial features, maintaining 100% likeness.",

"subject": {
"identity": "A stylish beautiful woman portrayed as Cleopatra, the eternal Queen of Egypt, radiating supreme authority, elegance, and divine power.",
"angle": "Full-body editorial portrait captured at a cinematic 3/4 angle, both Cleopatra and the horse positioned diagonally, rendered in ultra-crisp clarity with no blur.",
"pose": {
"body_position": "She is riding a majestic white horse, seated confidently with impeccable royal posture, her torso slightly turned to a 3/4 angle to emphasize grace and dominance.",
"hands": "One hand gently holds the gold-accented reins, while the other rests elegantly near her waist, displaying ornate jewelry.",
"expression": "She looks directly at the camera with a calm, commanding, and seductive gaze—the unmistakable presence of a queen born to rule."
}
},

"appearance": {
"outfit": "An opulent, ultra-bongga Cleopatra couture gown in pristine white and radiant gold. The gown features a sculpted corset bodice richly embroidered with gold hieroglyphic motifs, sun-disk patterns, and crystal beadwork. Flowing white silk, chiffon, and sheer organza panels cascade dramatically from the waist and shoulders, creating powerful movement as she rides. A daring thigh-high slit reveals her leg, balancing sensuality with imperial elegance. Every seam is traced with gold-thread embroidery for a luminous, goddess-like silhouette.",
"accessories": "She wears the exact Cleopatra headpiece from the attached reference image: a regal black-and-gold striped nemes-style headdress with a polished gold cobra (uraeus) centerpiece at the forehead, structured side panels, and intricate gold detailing. Paired with a wide Egyptian collar necklace with turquoise accents, engraved gold arm cuffs, crystal finger rings, an ornate gold waist belt, delicate anklets, and elegant flat Egyptian sandals.",
"hair": "Her hair is fully concealed beneath the nemes headpiece as shown in the reference image, ensuring perfect historical accuracy and symmetry.",
"makeup": "Ultra-bold, highly colorful Egyptian eye makeup with ceremonial ink artistry. Her eyes feature sharp elongated black kohl eyeliner extended dramatically past the outer corners, layered with vivid turquoise, teal, emerald green, sapphire blue, violet, and metallic gold pigments blended in high-fashion gradients. Beneath each eye, intricate hand-drawn ink designs inspired by ancient Egyptian symbolism are visible—fine black and gold lines, dots, and sacred motifs echoing hieroglyphs and protective markings, following the natural curve of the lower eye and cheekbone. Her complexion is flawless and softly bronzed with luminous highlights, brows are sculpted and powerful, cheeks carry a subtle coral-rose flush, and lips are finished in a refined nude-rose satin tone to balance the intense, artistic eye look."
},

"props": {
"animal": "A powerful, majestic white horse with a flowing ivory mane, sculpted muscles, and intelligent dark eyes. The horse wears elegant white-and-gold tack engraved with Egyptian motifs, symbolizing royal conquest, divine favor, and sovereignty."
},

"background": {
"macro_environment": "A vast open desert landscape near the Egyptian palace and pyramids at golden hour, stretching endlessly beneath a dramatic sky glowing with gold, amber, and soft ivory tones.",
"midground_details": "Distant pyramids rising from the sand, monumental stone statues, ceremonial banners moving in the wind, and faint silhouettes of royal guards and attendants placed far behind for scale.",
"micro_elements": "Fine desert sand lifted by the horse’s movement, sharply defined gold engravings on the reins, visible embroidery threads on the gown, subtle translucency of sheer fabrics, radiant light reflections on gold surfaces, and crisp, realistic shadows—everything rendered with extreme clarity and zero blur."
},

"lighting": {
"type": "Cinematic natural golden-hour lighting with soft reflective highlights.",
"effect": "Warm sunlight intensifies the white-and-gold palette, ignites the vibrant eye makeup colors and ink details, and illuminates the horse’s coat, creating a radiant, divine, editorial glow."
},

"camera": {
"camera_type": "DSLR",
"resolution": "8K UHD",
"lens": "50mm prime lens",
"aperture": "f/8 for maximum sharpness across subject and background",
"iso": 100,
"shutter_speed": "1/250s to freeze motion while maintaining realism",
"focus": "Extreme sharp focus from foreground to background, no bokeh, no blur"
},

"style": "High-fashion editorial, cinematic realism, divine Egyptian royalty, white-and-gold couture contrasted with vibrant ceremonial eye art, powerful, sensual, ultra-detailed, sharp, majestic"
} Prompt:
Use my image in Ultra-realistic, hyper-detailed, 8K cinematic portrait of a young stylish man, using the uploaded image for exact face and hairstyle.
Outfit: An oversized red knit sweater with white hearts, exactly as described in the prompt.
Pose: A hyper-realistic close-up portrait with a messy, cropped framing showing only the boy holding the book. His left hand rests on the wooden table and covers part of his cheek, with a subtle smile on his lips. His other hand holds the book titled "Something I Never Told You" with the word "YOU" written in pink, exactly as
described in the prompt. Background: Not specified.
滴滴出行優惠 👉 新用戶香港 Call 車首程免費(最高減 HK$88)— 按此領取優惠!