Text to image: Difference between revisions

From LemonWiki共筆
Jump to navigation Jump to search
 
(39 intermediate revisions by the same user not shown)
Line 2: Line 2:


{{Template:Generative AI Tool}}
{{Template:Generative AI Tool}}
{{LanguageSwitcher | content = [[Text to image | EN]], [[Text to image in Mandarin | 漢字]] }}


== List of AI image generators  ==
== List of AI image generators  ==
* {{Gd}} [https://leonardo.ai/ AI Art Generator - Create Art, Images & More | Leonardo AI] 150 daily free tokens, free to use
=== Online AI image generators ===
** Model: Leonardo Creative model or fine-tuned models<ref>[https://leonardo.ai/faq/ FAQ - AI Art Generator - Create Art, Images & More | Leonardo AI]</ref>
{{Gd}} [https://leonardo.ai/ AI Art Generator - Create Art, Images & More | Leonardo AI]  
* Free limit: 150 daily free tokens, free to use
* Model: Leonardo Creative model or fine-tuned models<ref>[https://leonardo.ai/faq/ FAQ - AI Art Generator - Create Art, Images & More | Leonardo AI]</ref>


* [https://beta.dreamstudio.ai/dream DreamStudio]
{{Gd}} [https://playgroundai.com/ Playground - free-to-use online AI image creator]  
* Free limit: Free to "create 500 images per day"
* Model: Stable Diffusion


* [https://colab.research.google.com/github/alembics/disco-diffusion/blob/main/Disco_Diffusion.ipynb Disco Diffusion v5.4]
{{Gd}} Google [https://gemini.google.com/app Gemini] (Formerly Google Bard) {{exclaim}} "It’s only available for English prompts"<ref>[https://support.google.com/gemini/answer/14286560?hl=en&co=GENIE.Platform%3DDesktop Generate images with Gemini Apps - Computer - Gemini Apps Help]</ref>
* Free limit: Free to use
* Model: [https://deepmind.google/technologies/imagen-2/ Imagen 2 - Google DeepMind]  


* [https://playgroundai.com/ Playground - free-to-use online AI image creator] free to use {{Gd}}
{{Gd}} [https://huggingface.co/spaces/radames/Real-Time-Text-to-Image-SDXL-Lightning Real-Time Text-to-Image SDXL Lightning - a Hugging Face Space by radames]
** Model: Stable Diffusion
* Free limit: Free to use. Login is not required.
* Model: [https://huggingface.co/ByteDance/SDXL-Lightning ByteDance/SDXL-Lightning · Hugging Face]


* [https://www.midjourney.com/app/ midjourney]
{{Gd}} [https://www.bing.com/images/create Microsoft Designer] or [https://designer.microsoft.com/home Microsoft Designer - Stunning designs in a flash]
** Community: [https://www.facebook.com/groups/midjourneytw/ Midjourney AI 台灣社群 | Facebook]
* Free limit: Free to use. Login is required.
** Documentation: [https://midjourney.gitbook.io/docs/ Quick Start Guide - Midjourney Documentation]
* Model: DALL-E 3<ref>[https://designer.microsoft.com/FAQ.pdf FAQ.pdf]</ref>
* Example prompts: [[Prompts of Bing image creator]]


* OpenAI.com [https://chat.openai.com/ ChatGPT]
[https://www.craiyon.com/ Craiyon - Your FREE AI image generator tool: Create AI art!]
** Model: [https://openai.com/dall-e-3 DALL·E 3]
* Free limit:  free to use; About 50+ seconds for 9 images. Login is not required.
* Model: [https://github.com/borisdayma/dalle-mini borisdayma/dalle-mini: DALL·E Mini - Generate images from a text prompt]: [https://huggingface.co/spaces/dalle-mini/dalle-mini online demo] created by Boris Dayma


* [https://github.com/borisdayma/dalle-mini borisdayma/dalle-mini: DALL·E Mini - Generate images from a text prompt]: [https://huggingface.co/spaces/dalle-mini/dalle-mini online demo] created by Boris Dayma. free to use {{Gd}}
[https://tensor.art/ Tensor.Art] {{exclaim}} {{Acronym | acronym=NSFW | def=上班不要看、Not Suitable For Work}}
** Model: VQGAN-f16-16384 & DALL·E mini or DALL·E mega<ref>[https://github.com/borisdayma/dalle-mini borisdayma/dalle-mini: DALL·E Mini - Generate images from a text prompt]</ref>
* Model: [https://tensor.art/models/619225630271212879 Stable Diffusion XL(SDXL)1.0], import model from [https://civitai.com/ Civitai: The Home of Open-Source Generative AI]
* Free limit: 100 free images daily {{Gd}}


* Google
[https://creator.nightcafe.studio/studio NightCafe Creator] {{access | date=2024-03-03}}
** ❌ [https://bard.google.com/chat Bard] not suppoted
* Free limit: Free to "create 10 images per day"
** [https://imagen.research.google/ Imagen: Text-to-Image Diffusion Models]
* Model: Stable Diffusion


* [https://www.canva.com/your-apps/text-to-image-(beta) Texy to image 發掘應用程式 - Canva]
[https://stablediffusionweb.com/zh-tw Stable Diffusion 在線]
** Terms: [https://www.canva.com/policies/text-to-image-terms/ Text to Image Terms]
* Free limit: Free to "create 10 images per day"
* Model: Stable Diffusion


* [https://www.adobe.com/tw/products/firefly.html Adobe Firefly – 適合所有人的生成式 AI]
[https://replicate.com/black-forest-labs/flux-pro black-forest-labs/flux-pro flux-pro] ([https://blog.fal.ai/flux-the-largest-open-sourced-text2img-model-now-available-on-fal/ Announcing Flux by Black Forest Labs: The Next Leap in Text-to-Image Models])
* Free limit:
* Model:


== Open Source Models ==
[https://beta.dreamstudio.ai/dream DreamStudio]
* [https://huggingface.co/spaces/stabilityai/stable-diffusion Stable Diffusion - a Hugging Face Space by stabilityai]
* Model
** Source of training data: [https://laion.ai/ LAION]<ref>[https://waxy.org/2022/08/exploring-12-million-of-the-images-used-to-train-stable-diffusions-image-generator/?fbclid=IwAR12dm-9_qe_pIoeR6PZQx1JQdmRrEJ3QsMjgU1SvSmPxeVxj4Tj47NuFAc Exploring 12 Million of the 2.3 Billion Images Used to Train Stable Diffusion's Image Generator - Waxy.org]</ref>
 
[https://colab.research.google.com/github/alembics/disco-diffusion/blob/main/Disco_Diffusion.ipynb Disco Diffusion v5.4]
* Model
 
[https://www.midjourney.com/app/ midjourney]
* Community: [https://www.facebook.com/groups/midjourneytw/ Midjourney AI 台灣社群 | Facebook]
* Documentation: [https://midjourney.gitbook.io/docs/ Quick Start Guide - Midjourney Documentation]
 
[https://www.rubbrband.com/home rubbrband]
* Free Limit: Rubbrband offers a free trial, with specific usage limits and terms available on its official pricing page.
* Model: Rubbrband supports multiple AI models, including Stable Diffusion, DALL-E, and PixArt, allowing users flexible options to create various styles of images.
* Features: Custom Workflows, Multi-Model Support, Prompt Engineering Assistance, Color Control & AI Enhancement<ref>[https://aipure.ai/products/rubbrband/features?utm_source=chatgpt.com Rubbrband Features, Pros, Cons, and Use Cases]</ref>
 
[https://chendaryen.github.io/NitroFusion.github.io/ NitroDiffusion] (Demo: [https://huggingface.co/spaces/ChenDY/NitroFusion_1step_T2I NitroFusion 1step T2I - a Hugging Face Space by ChenDY])
* Free Limit:
* Model: "The model, NitroFusion, represents a world first and has been made open source by its developers – SketchX, a lab within PAI – a move that fundamentally transforms access to AI-enabled image creation models for creative professionals."<ref>[https://www.surrey.ac.uk/news/surrey-announces-worlds-first-ai-model-near-instant-image-creation-consumer-grade-hardware Surrey announces world's first AI model for near-instant image creation on consumer-grade hardware | University of Surrey]</ref>
* Maintainer: Surrey Institute for People-Centred Artificial Intelligence (PAI) at the University of Surrey
 
''$'' OpenAI.com [https://chat.openai.com/ ChatGPT]
* Model: [https://openai.com/dall-e-3 DALL·E 3] ([[Troubleshooting of OpenAI DALLE3 | Troubleshooting of OpenAI DALLE-3]])
 
Google [https://aitestkitchen.withgoogle.com/tools/image-fx ImageFX] {{exclaim}} Not available in Taiwan
* Model: [https://deepmind.google/technologies/imagen-2/ Imagen 2 - Google DeepMind] (formerly [https://imagen.research.google/ Imagen: Text-to-Image Diffusion Models])


== Image to text prompt ==
[https://www.canva.com/your-apps/text-to-image-(beta) Texy to image 發掘應用程式 - Canva]
* [https://colab.research.google.com/github/pharmapsychotic/clip-interrogator/blob/main/clip_interrogator.ipynb#scrollTo=OGmvkzITN4Hz clip_interrogator.ipynb - Colaboratory]
* Terms: [https://www.canva.com/policies/text-to-image-terms/ Text to Image Terms]
* Model:


== Text prompt to mockup diagram ==
[https://www.adobe.com/tw/products/firefly.html Adobe Firefly – 適合所有人的生成式 AI]
* [https://github.com/tldraw/make-real-starter tldraw/make-real-starter: Make it real] powered by DALL-E
* Model:


== Software ==
=== Software of AI image generators ===


* {{Mac}} [https://diffusionbee.com/ DiffusionBee - Stable Diffusion App for AI Art] ([https://github.com/divamgupta/diffusionbee-stable-diffusion-ui github]) on {{Mac}} "Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed." {{exclaim}} Prompt in Chinese was not supported on Version 1.5.1（0016）.
* {{Mac}} [https://diffusionbee.com/ DiffusionBee - Stable Diffusion App for AI Art] ([https://github.com/divamgupta/diffusionbee-stable-diffusion-ui github]) on {{Mac}} "Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed." {{exclaim}} Prompt in Chinese was not supported on Version 1.5.1（0016）.
Line 56: Line 94:
** Models: Stable Diffusion v. 1.5 NOT v. 2
** Models: Stable Diffusion v. 1.5 NOT v. 2
** Using custom models:
** Using custom models:
* {{Mac}} [https://apps.apple.com/us/app/draw-things-ai-generation/id6444050820 Draw Things: AI Generation]
** Models: SDXL, Stable Diffusion ...


* {{Linux}} [https://github.com/AUTOMATIC1111/stable-diffusion-webui AUTOMATIC1111/stable-diffusion-webui: Stable Diffusion web UI]
* {{Linux}} [https://github.com/AUTOMATIC1111/stable-diffusion-webui AUTOMATIC1111/stable-diffusion-webui: Stable Diffusion web UI]
Line 64: Line 105:
** Models: Stable Diffusion
** Models: Stable Diffusion
** Using custom models:
** Using custom models:
== Open Source Models ==
* [https://huggingface.co/spaces/stabilityai/stable-diffusion Stable Diffusion - a Hugging Face Space by stabilityai]
** Source of training data: [https://laion.ai/ LAION]<ref>[https://waxy.org/2022/08/exploring-12-million-of-the-images-used-to-train-stable-diffusions-image-generator/?fbclid=IwAR12dm-9_qe_pIoeR6PZQx1JQdmRrEJ3QsMjgU1SvSmPxeVxj4Tj47NuFAc Exploring 12 Million of the 2.3 Billion Images Used to Train Stable Diffusion's Image Generator - Waxy.org]</ref>
== Image to text prompt ==
* [https://colab.research.google.com/github/pharmapsychotic/clip-interrogator/blob/main/clip_interrogator.ipynb#scrollTo=OGmvkzITN4Hz clip_interrogator.ipynb - Colaboratory]
=== Prompt ideas ===
* [https://lexica.art/ Lexica]: "The Stable Diffusion search engine"
* [https://arthub.ai/ Arthub.ai]: "Discover, Upload and Share AI Generated Art"
* [https://stablediffusionweb.com/zh-tw/prompts 搜索提示 - 為Stable Diffusion創建更好的提示 | Stable Diffusion 在線]
* [https://prompthero.com/ PromptHero - Search prompts for Stable Diffusion, ChatGPT & Midjourney]
* [https://freeflo.ai/ AI image prompts for Midjourney, Dall-E, Stable Diffusion]
* [https://www.bing.com/images/create Image Creator from Microsoft Designer]
== Text prompt to mockup diagram ==
* [https://github.com/tldraw/make-real-starter tldraw/make-real-starter: Make it real] powered by DALL-E
== Logo generator ==
* [https://ideogram.ai/t/explore Ideogram]
* ''$'' [https://logodiffusion.com/ Create Logos in Seconds With Generative A.I | Logo Diffusion]
== Image to image ==
[https://magnific.ai/ Magnific AI — The magic image Upscaler & Enhancer]
* How Does Magnific Work?: Transform any image into a higher resolution version, precisely adding the details you need. At Magnific
* Applicable Users: Professional photographers, graphic designers, digital artists, and illustrators
[https://getimg.ai/image-editor Image Editor | getimg.ai]
* [https://getimg.ai/guides/working-with-and-organizing-generated-images Working with and organizing generated images | getimg.ai]


== Community ==
== Community ==
Line 70: Line 141:
* [https://www.facebook.com/groups/midjourneybn/ Midjourney AI 鳥巢 | Facebook]
* [https://www.facebook.com/groups/midjourneybn/ Midjourney AI 鳥巢 | Facebook]
* [https://www.facebook.com/groups/officialmidjourney Midjourney Official | Facebook]
* [https://www.facebook.com/groups/officialmidjourney Midjourney Official | Facebook]
* [https://www.facebook.com/groups/592850912909945 Bing DALL-E 3 and ALL AI 生成式藝術小小詠唱師 | Facebook]
* [https://www.facebook.com/groups/sdaitw Stable Diffusion 台灣社群v2 | Facebook]


== Prompt ideas ==
== Commercial usage ==
* [https://lexica.art/ Lexica]: "The Stable Diffusion search engine"
* Midjourney: "If you have subscribed at any point, you are free to use your images in just about any way you want. You must purchase the Pro or Mega plan if you are a company making more than $1,000,000 USD in gross revenue per year. For complete details, please see the [https://docs.midjourney.com/terms-of-service Terms of Service]" <ref>[https://docs.midjourney.com/docs/plans Midjourney Subscription Plans]</ref> 「如果你曾經訂閱過，你可以幾乎以任何方式自由使用你的圖片。如果你是一家年營業額超過100萬美元的公司，你必須購買專業版或超級版方案。欲了解完整詳情，請參閱[https://docs.midjourney.com/docs/plans 服務條款]。」
* [https://arthub.ai/ Arthub.ai]: "Discover, Upload and Share AI Generated Art"


== References ==
== References ==
<references />
<references />


== Further reading ==
== Further reading ==

Latest revision as of 15:51, 11 December 2024

Creates an image from the text description.

Generative AI Tool: 📁 Prompts | 👩🏻‍💻 Text to presentation | 📢 Text to audio or speech | 🎵 Text to music | 🖼️ Text to image | 🎥 Text to video | 🚀 Coding | 🧭 Guideline | 💬 FAQ


🌐 Switch language: EN, 漢字


List of AI image generators[edit]

Online AI image generators[edit]

Good.gif AI Art Generator - Create Art, Images & More | Leonardo AI

  • Free limit: 150 daily free tokens, free to use
  • Model: Leonardo Creative model or fine-tuned models[1]

Good.gif Playground - free-to-use online AI image creator

  • Free limit: Free to "create 500 images per day"
  • Model: Stable Diffusion

Good.gif Google Gemini (Formerly Google Bard) Icon_exclaim.gif "It’s only available for English prompts"[2]

Good.gif Real-Time Text-to-Image SDXL Lightning - a Hugging Face Space by radames

Good.gif Microsoft Designer or Microsoft Designer - Stunning designs in a flash

Craiyon - Your FREE AI image generator tool: Create AI art!

Tensor.Art Icon_exclaim.gif NSFW

NightCafe Creator [Last visited: 2024-03-03]

  • Free limit: Free to "create 10 images per day"
  • Model: Stable Diffusion

Stable Diffusion 在線

  • Free limit: Free to "create 10 images per day"
  • Model: Stable Diffusion

black-forest-labs/flux-pro flux-pro (Announcing Flux by Black Forest Labs: The Next Leap in Text-to-Image Models)

  • Free limit:
  • Model:

DreamStudio

  • Model

Disco Diffusion v5.4

  • Model

midjourney

rubbrband

  • Free Limit: Rubbrband offers a free trial, with specific usage limits and terms available on its official pricing page.
  • Model: Rubbrband supports multiple AI models, including Stable Diffusion, DALL-E, and PixArt, allowing users flexible options to create various styles of images.
  • Features: Custom Workflows, Multi-Model Support, Prompt Engineering Assistance, Color Control & AI Enhancement[4]

NitroDiffusion (Demo: NitroFusion 1step T2I - a Hugging Face Space by ChenDY)

  • Free Limit:
  • Model: "The model, NitroFusion, represents a world first and has been made open source by its developers – SketchX, a lab within PAI – a move that fundamentally transforms access to AI-enabled image creation models for creative professionals."[5]
  • Maintainer: Surrey Institute for People-Centred Artificial Intelligence (PAI) at the University of Surrey

$ OpenAI.com ChatGPT

Google ImageFX Icon_exclaim.gif Not available in Taiwan

Texy to image 發掘應用程式 - Canva

Adobe Firefly – 適合所有人的生成式 AI

  • Model:

Software of AI image generators[edit]

  • macOS icon_os_mac.png DiffusionBee - Stable Diffusion App for AI Art (github) on macOS icon_os_mac.png "Diffusion Bee is the easiest way to run Stable Diffusion locally on your M1 Mac. Comes with a one-click installer. No dependencies or technical knowledge needed." Icon_exclaim.gif Prompt in Chinese was not supported on Version 1.5.1（0016）.
    • Models: Stable Diffusion
    • Using custom models: Allowed[6]
  • macOS icon_os_mac.png Diffusers by Hugging Face
    • Models: Stable Diffusion 1.4 ~ 2.1 & OFA small v0
    • Using custom models:
  • macOS icon_os_mac.png Amazing AI "Text-to-Image Stable Diffusion" (Amazing AI - Acknowledgements) Icon_exclaim.gif Prompt in Chinese was not supported on Version 1.0.3 (6).
    • Models: Stable Diffusion v. 1.5 NOT v. 2
    • Using custom models:

Open Source Models[edit]

Image to text prompt[edit]

Prompt ideas[edit]

Text prompt to mockup diagram[edit]

Logo generator[edit]

Image to image[edit]

Magnific AI — The magic image Upscaler & Enhancer

  • How Does Magnific Work?: Transform any image into a higher resolution version, precisely adding the details you need. At Magnific
  • Applicable Users: Professional photographers, graphic designers, digital artists, and illustrators

Image Editor | getimg.ai

Community[edit]

Commercial usage[edit]

  • Midjourney: "If you have subscribed at any point, you are free to use your images in just about any way you want. You must purchase the Pro or Mega plan if you are a company making more than $1,000,000 USD in gross revenue per year. For complete details, please see the Terms of Service" [8] 「如果你曾經訂閱過，你可以幾乎以任何方式自由使用你的圖片。如果你是一家年營業額超過100萬美元的公司，你必須購買專業版或超級版方案。欲了解完整詳情，請參閱服務條款。」

References[edit]

  1. ↑ FAQ - AI Art Generator - Create Art, Images & More | Leonardo AI
  2. ↑ Generate images with Gemini Apps - Computer - Gemini Apps Help
  3. ↑ FAQ.pdf
  4. ↑ Rubbrband Features, Pros, Cons, and Use Cases
  5. ↑ Surrey announces world's first AI model for near-instant image creation on consumer-grade hardware | University of Surrey
  6. ↑ Using custom models
  7. ↑ Exploring 12 Million of the 2.3 Billion Images Used to Train Stable Diffusion's Image Generator - Waxy.org
  8. ↑ Midjourney Subscription Plans

Further reading[edit]

Retrieved from "https://wiki.planetoid.info/index.php?title=Text_to_image&oldid=25230"