How does Promptsref help me write better AI prompts?

Promptsref connects examples, prompts, search, and creation tools in one workflow. You can search for a style, image, video, or prompt, adjust the subject, style, and details, then test the direction with our generation tools.

Can I generate images or videos here directly?

Yes. Pick a reference, open its prompt, change the subject, ratio, style, or model, and create a new version with our AI tools. After generation, you can keep editing and compare different directions quickly.

Is Promptsref only for Midjourney?

No. Promptsref started with SREF research, but it now supports broader AI image and video workflows. You can browse the prompt library, compare model examples, and continue creating with the tools on the site.

Why not just ask an AI chatbot to write a prompt?

You can, but many AI chatbot answers are generic. Promptsref shows you the result first and then the prompt behind it, which makes the creative direction more concrete: style, composition, lighting, and useful details.

Who is Promptsref for?

Promptsref is useful for designers, marketers, video creators, and independent creators. It helps when you want faster AI results without losing control, with reliable prompts, clear references, and quick variations.

Curated AI promptsand generation toolsMake creating simple

Explore Today’s Picks Start Generating

★★★★★Rated 5.0

Loved by 10k+ creators worldwide

All Models in One Place

10K+ Public Prompts

One-Click Trending AI Effect Videos

Generate a Photo

Generate Viral Videos

AI Change Angle

Remove Background

Tool library

AI Image Generator

Generate images from text prompts using multiple AI models simultaneously

AI Effect Templates

Image to Prompt Generator

Upload a reference image and turn it into detailed JSON and natural-language prompts.

Upscale Image AI 4K

Upscale images to 2K, 4K, or 8K while preserving important details.

Change Camera Angle

Generate new camera angles from a single photo

Crop & Remove Background

Crop images and remove backgrounds.

Split Image Into Grid

Split an image into smaller pieces and upscale each tile to HD quality

Generate Video

Generate videos from text or keyframes using multiple AI video models in a single workflow.

Generate Voice

Create natural AI voiceovers with text to speech, voice clone, and voice design using Qwen3 TTS.

Video Motion Control

Generate videos with motion control using AI.

Free Video Frame Extractor

Extract frames from videos instantly.

Free Turn Grid to Video

Upload grid images and convert them to videos.

Text to Speech

Turn scripts into natural voiceovers.

Voice Clone

Use a permitted reference audio to match a voice.

Voice Design

Describe a custom speaker and delivery style.

Explore Prompt Inspirations

All

GPT Image

Nano Banana

Seedance 2.0

Image generated by gpt-image/2.0-text-to-image, created by xie underwood

{ "subject": { "description": "A clearly adult young woman crouches low beside an outdoor curb, her compact body cent...

Premium editorial campaign. Main Subject: Young male model wearing a vivid turquoise utility jacket. Pose: Both hands...

Image generated by Midjourney, created by Promptsref SREF

Image generated by gpt-image/2.0-text-to-image, created by Kyoto Lust

generate a twitter banner with the dimensions of 1500 x 500, a sleek and dark ethereal vibe with an ominous old Engli...

{ "subject": { "description": "A young adult woman seated beside a café table, taking a close-range smartphone selfie...

8204

Image generated by midjourney, created by xie underwood

하늘이 떠다니는 물고기 A high tech high-fashion editorial photograph of a woman robot flying on a high tech flyer saucer throug...

233

Create an authentic 3:4 vertical smartphone photograph taken at an outdoor urban café. A young adult woman, approxima...

11268

Image generated by gpt-image/2.0-text-to-image, created by yuechen ye

设计一个微信小程序「轰趴馆」活动中心页面，整体采用传统本地生活团购风格，参考美团、大众点评、抖音团购的设计语言。界面简洁、干净、现代，白色背景搭配浅灰色页面底色，卡片式布局，大面积留白，圆角 16px，轻微阴影，突出信息层级和易读性。...

Create a clean 9:16 vertical smartphone portrait with absolutely no social-media UI, status bar, text, captions, icon...

046

Image generated by gpt-image/2.0-text-to-image, created by 司波雪奏

@創作圖像 VTuber 基礎設定藝名：深雪．冰華 (Miyuki Hyouka)稱號：四葉家的絕對冰后 / 現代魔法的頂點歌姬身高：160cm繪師（P站老爹）：石田可奈（風格致敬）粉絲名：CAD齒輪 / 雪結晶🎨 外觀與服裝設計髮...

628

Image generated by gpt-image/2.0-text-to-image, created by Patrick Lee (Zyceson)

鸣潮心月狐成年角色真人COS，极简渐变背景高端概念人像，极简时尚封面感 × 高级 editorial 人像 × 独立艺术企划主视觉。角色保留原本核心识别点，包括发型、发色、瞳色、服装轮廓、配饰、代表性色彩与气质，并自动转化为更适合真...

253

Image generated by gpt-image/2.0-text-to-image, created by 林浩强

娜美 Nami（海贼王），成年角色真人 COS，真实摄影质感，高级商业摄影风格。保留娜美核心角色识别点，包括橙色长发、棕色眼睛、成熟自信的航海士气质、草帽海贼团成员身份、性感灵动的女性魅力、航海服装风格、天候棒武器、航海地图元素、天...

7130

Use the uploaded face as the exact facial reference.A premium 9:16 streetwear editorial advertising poster.Main Subje...

012

AI Effect

Image generated by gpt-image/2.0-text-to-image, created by Sam Malik

A photo of Grace Ashcroft high-sport fashion editorial campaign photograph,white clear backgrund 35mm film, subtle gr...

4101

Image generated by gpt-image/2.0-text-to-image, created by 林言澈

凯多成年真人COS，真实摄影质感。保留角色核心识别点：发型、服装配色、武器、能力属性、身份特征、阵营符号与世界观设定。转译为真实可穿戴高级COS服装，融合护甲、金属部件、饰品、专属装备。 9:16竖版构图。人物坐在古典石柱或未来...

022

Premium editorial campaign. Main Subject: Male model wearing a vivid electric-blue hoodie. Pose: Slight forward lean ...

Image generated by gpt-image/2.0-text-to-image, created by Penn OG

生成一张 9:16 竖版高级科幻机甲电影角色海报，真人电影质感，院线 IMAX 主视觉风格。角色是一位银白色凌乱短发的年轻男性机甲驾驶员，皮肤苍白细腻，五官清冷俊美，眼神孤独、温柔、坚定，微微仰头望向远方。右上方占据画面大面积的是他...

027

Image generated by gpt-image/2.0-text-to-image, created by Matthew Thias

A image of a teenager walking on air with text saying BRO HAS INFINITE RIZZ FOR EONS

015

创建一张 8K 超高清、电影级质感的广角壁纸，采用三分法构图，两位《咒术回战》角色面对面展开一场充满张力的终极对峙。左侧为五条悟，身穿黑色高领制服，可佩戴眼罩或开启六眼状态，周身散发耀眼的蓝白色咒力与空间能量，神情冷静而充满压迫感；右...

237

卡塔库栗成年真人COS，真实摄影质感。保留角色核心识别点：发型、服装配色、武器、能力属性、身份特征、阵营符号与世界观设定。转译为真实可穿戴高级COS服装，融合护甲、金属部件、饰品、专属装备。 9:16竖版构图。人物坐在古典石柱或...

030

9:16 vertical fashion editorial photography, photorealistic, top-down overhead shot, lying on lush green grass at nig...

Image generated by gpt-image/2.0-text-to-image, created by Sheeraz “Kat” Aman

toyota mk4 dodge demon srt hemi swapped

{ "subject": { "description": "An adult woman around 20 to 27 years old sits near the edge of a bed covered in layere...

8160

Image generated by gpt-image/2.0-text-to-image, created by Samuel Brown

Make an owl threatening a geometry dash cube with an owl gun.

Image generated by gpt-image/2.0-text-to-image, created by 王Jason

<role> You are an award-winning trailer director, cinematographer, and storyboard artist. Your job: turn ONE referenc...

136

强尼·银手成年真人COS，真实摄影质感。保留角色核心识别点：发型、服装配色、武器、能力属性、身份特征、阵营符号与世界观设定。转译为真实可穿戴高级COS服装，融合护甲、金属部件、饰品、专属装备。 9:16竖版构图。人物坐在古典石柱...

132

Subject: Premium Sneakers Appearance: Premium white and black sneakers with breathable mesh, smooth leather finish, c...

{ "style_rules": { "background": "crimson red + ice-blue diagonal block", "character": "black silhouette, side profil...

025

Create a realistic 3:4 vertical environmental portrait with the feeling of an intimate everyday snapshot. An adult wo...

10131

Image generated by gpt-image/2.0-text-to-image, created by 唐嘉艺

[Overall Composition] A horizontal panoramic view, focusing the visual attention on the giant cheese arch bridge in t...

128

高品质原创二次元游戏角色插画，商业游戏宣传原画，日系板绘风格，现代军事美学，战术少女，日本现代特种作战风格，日本自卫队特殊作战人员风格，原创角色设计，年轻成年女性，八头身比例，身高约168cm，身材修长紧实，偏敏捷型体格，肩部线条自然...

331

鱼眼广角深夜街头海报，16:9 横版——采用极限运动摄影机级别的鱼眼广角镜头，贴地仰拍构图，画面带明显桶形畸变，人物微微前倾像要探出画面；中央是[角色搭配]，二次元厚涂画风，睡眼惺忪、脸颊泛红晕染，抬头直视镜头；背景是一间真实感的[店...

217

Frequently Asked Questions

Short answers for creators who want to start faster.

Promptsref is an AI workspace for creators. It collects curated images and videos together with the prompts behind them, so you can find a result you like, study how it was written, and create with our AI tools without starting from a blank page.