#text2image
I could be wrong. Do you know anything about using AI other than "You use prompts"? Any tools? Any techniques? Anything other than basic text2image prompting through an LLM?
November 8, 2026 at 4:08 AM
combined with kaibo's tts, text2image, and image2image, in a moderately smart agent harness, it can crank out pretty much everything on the fly
November 1, 2026 at 1:56 PM
#ai #aiart #aigirl #duck.ai #online #sfw #retrofuturism #spacegirl #text2image #borrowedprompt
found the prompt in the alt-text in a QP by @majorbill.bsky.social copy/pasted into duck.ai with a style tweak. I need to see if it will let me control the width and height too.
October 17, 2026 at 9:46 PM
It wasn't made to look good. It's a very basic setup to proof deterministic processes. Simple prompt, straightforward sampler, no extra tools or weight modifiers. Just... a basic text2image workflow. Which should be obvious, since it's all literally right there on screen for you to see.
October 4, 2026 at 9:48 PM
it is completely rational to oppose "the good parts" of ai, if you believe "the bad parts" outweigh it, because you can't have one without the other. the distinction between "ai" and "genai" that people make doesn't actually exist under the hood. text2image was born from google reverse image search
August 27, 2026 at 2:47 AM
[2/2] the text to image leaderboard of Artificial Analysis, an independent AI model evaluation and analysis platform. Its score is second only to OpenAI and surpasses Google Nano Banana 2 (Gemini 3.1 Flash Image Preview), NVIDIA Cosmos3-Super-Text2Image and ByteDance Seedream 4.0, etc.
June 10, 2026 at 2:10 PM
[2/2] the text to image leaderboard of Artificial Analysis, an independent AI model evaluation and analysis platform. Its score is second only to OpenAI and surpasses Google Nano Banana 2 (Gemini 3.1 Flash Image Preview), NVIDIA Cosmos3-Super-Text2Image and ByteDance Seedream 4.0, etc.
June 10, 2026 at 2:10 PM
[2/2] the text to image leaderboard of Artificial Analysis, an independent AI model evaluation and analysis platform. Its score is second only to OpenAI and surpasses Google Nano Banana 2 (Gemini 3.1 Flash Image Preview), NVIDIA Cosmos3-Super-Text2Image and ByteDance Seedream 4.0, etc.
June 10, 2026 at 2:10 PM
[2/2] the text to image leaderboard of Artificial Analysis, an independent AI model evaluation and analysis platform. Its score is second only to OpenAI and surpasses Google Nano Banana 2 (Gemini 3.1 Flash Image Preview), NVIDIA Cosmos3-Super-Text2Image and ByteDance Seedream 4.0, etc.
June 10, 2026 at 2:03 PM
Open Image Preferences released! 🚀

- Open-source dataset for text2image
- 10K samples manually evaluated by the HF community.
- Binarized format for SFT, DPO, or ORPO.

It comes with a nice blog post explaining the steps to pre-process and generate the data, along with the results.
December 9, 2024 at 4:26 PM
+ Up to 231% improvement over other bench-marked models (see blog below).
+ Evaluated on over 4M products across 10,000's of categories.
+ Detailed performance comparisons across three major tasks: Text2Image, Category2Image, and AmazonProducts-Text2Image.
November 24, 2024 at 10:01 PM
🤖🎨 Zoo

Con todas sus limitaciones la respuesta de los modelos AI text2image son espectaculares…

Puedes probarlos aquí
https://zoo.replicate.dev
August 16, 2023 at 5:22 PM
Os acerco Zoo, un servicio online con el que podrás probar gratis (desde el móvil) hasta 8 modelos generativos text2image.

SDXL
Stablediffusion 2.1
Stable-diffusion 1.5
Kandinsky-2.2
Kandinsky-2
DALLE2
Deepfloyd-if
Material-diffusion

https://zoo.replicate.dev/
August 5, 2023 at 6:35 AM
Any comments on the AI image being made using only text2image?

The AI gen metadata shows an image was not used for img2img. It can be reproduced on another machine using same inputs.

So this AI image coincidentally happens to have all the lines and details in the same place?
October 16, 2025 at 1:40 PM
allkeyai will soon support Seedance 2.0! If you're interested, feel free to give it a try.
allkeyai:https://allkey.ai/console/text2image/?invite=1a12
February 26, 2026 at 12:38 AM
Andererseits vermutlich auch genau das Ausmaß an Kreativität, das das durchschnittliche Publikum mit diesen Tools bewerkstelligen wird. Wer einmal angeschaut hat, wie unfassbar fantasielos der Output von Text2Image-AI ist, weiß das. "Cyberpunk Anime Girl with cool dragon", say no more
September 18, 2024 at 8:57 AM
Die tatsächlich bislang spürbare "kreative Zerstörung" durch die ganzen tollen Techdisruptoren wie Über, AirBnB, LLMs und Text2Image-AIs beschränkt sich auf die originell gesetzwidrige Umgehung von Konsumenten-, Mieter- und Arbeitnehmerrechten auf Kosten der Allgemeinheit
January 7, 2024 at 3:41 PM
Haunted Liminal Places... Maps, fake places, and more...

arnicas.substack.com/p/titaa-32-h...
TITAA #32: Haunted Liminal Places
Phantom Settlements - Custom Text2Image Models - Zooming - Crazed CEOs
arnicas.substack.com
February 11, 2025 at 5:05 AM