home.social

#text2image — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #text2image, aggregated by home.social.

fetched live
  1. Introducing ChatGPT Images, powered by our flagship new image generation model.

    - Stronger instruction following
    - Precise editing
    - Detail preservation
    - 4x faster than before

    Rolling out today in ChatGPT for all users, and in the API as GPT Image 1.5.

    x.com/openai/status/2000990989

    @_OpenAI #AI #Text2Image #OpenAI

  2. There's an ongoing drive for #FOSS #distributedAI computing, e.g. #aiHorde. It's not perfect & needs a 3rd-party security audit, but is somewhere on the path to more-secure non-local #edgecomputing for devices like phones that can't load large #AImodels or run popular #text2text / #text2image #genAI applications locally.

    *NB: Apologies to Horde volunteers if I just bumped your traffic past reasonable limits.*

    aihorde.net

  3. There's an ongoing drive to create #FOSS open source, #distibutedAI computing. #AI_Horde is one such project, and possibly the most well-known right now. It's not perfect, and no one I know has done an in-depth security analysis, but I suspect this is somewhere on the path to safer non-local AI computing for edge devices like phones that simply can't run larger #AImodels or popular #text2text or #text2image #genAI applications locally.

    NB: My apologies to the Horde volunteers if I just bumped your traffic past reasonable limits.

    aihorde.net

  4. - Nova Premier (Q1 2025):
    • Ultimate #MachineLearning capabilities
    • Advanced #ModelDistillation
    • Complex #ReasoningAI tasks
    #TeacherModel functionality

    🎨 Creative Suite #AIGC #GenerativeAI:
    - #NovaCanvas:
    #Text2Image generation

  5. #Text2Image with large models is both kinda boring and getting better by the minute. What is *not* boring is fine tuning larger models with smaller datasets for more nuanced output. Way more practical. Today I tuned weights on the #Flux1 model. Did not disappoint...

    1/7

  6. Discussion about #generativeAI is also about it's requirements in terms of compute. But big progress is made in running models on smaller platforms, like this #text2image model, #mobilediffusion that runs on high-end android and iOS devices.
    blog.research.google/2024/01/m

  7. lo-fi Shepard tone albums to flick and careen to

    youtube.com/watch?v=hWGNpSmvWt — TOP 500 AI GENERATED ART #text2image (1:32:18)

  8. Tewel et al: Key-Locked Rank One Editing for Text-to-Image Personalization.
    A 100kb only text2image model #perfusion trained in only 4 minutes by NVIDIA. SIGGRAPH2023 paper. I hope they are going to publish the code for the conference...
    Paper: research.nvidia.com/labs/par/P
    #creativeAI #generativeAI #AI #artificialintelligence #text2image #siggraph

  9. I find these timelines very useful. Only that for me the interesting starts with Parti and Midjourney v4 (there's also a thing called Structured Diffusion Guidance), because these models also work on scene composition. Everything before that is pure randomness #text2image #AIart

    Image from arxiv.org/abs/2307.07521v1

  10. A recent paper proposes a systematic approach for prompt refining for #text2image generators. Yet, they work only on style. No consideration of scene composition, which some generators now support. #AIart

    creative.ai/@arxiv_hc/11074843

  11. #A #Text2Image
    A research tool that can generate panoramic 360º skybox prototypes from a few words. It allows users to save the skybox to their local devices, share a link to the skybox, and join a Discord channel to provide feedback and qualify for early access to new features. The tool is free to use

    skybox.blockadelabs.com

  12. CW: Made by openAI - Playing with the dragon

    Playing with the dragon - Meine Tochte wünschte sich von DallE2 openAi eine Bild von Mädchen das mit einem lieben Drachen spielt. Sie fand das Ergebnis toll und ich mag es auch :)
    #openai #DALLE2 #text2image #art #digitalArt #Mastoart

  13. #BlueWillow Vs. #MidJourney

    Angeblich soll BlueWillow ein Midjourney killer sein. Bin gerade der Beta beigetreten und starte den Vergleich… auch zurm lokal laufenden Automatic1111.

    Kommt mur wie ein KI Wettrüsten vor 😅
    #ai #StableDiffusion #text2image #machinelearning

    bluewillow.ai/

  14. The Giger Dragon - Ein #AIGenersted #text2image Versuch den ich gar nicht mal so schlecht finde.

  15. Seit ich gestern #StableDiffusion kostenlos auf meinem eigenen Rechner nutzen kann, rennt mein Rechner durchgängig durch hunderte prompts. Noch fallen knapp 65% der Images durch mein Qualitätsnetz, aber es wird besser und besser.
    Die Performance ist großartig. Aufrüsten hat sich in dem Fall mal gelohnt :)
    #aigenerated #Art #text2image

  16. Would you play your #vinyl records on this device? And what would you think they'd sound like? #text2image #stablediffusion #aiimage