home.social

#flux1 — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #flux1, aggregated by home.social.

  1. CW: Arte de IA, AI Art

    Garça de magia azul sobre floresta de bambus. Partículas de energia pura nas asas e cauda.

    Blue magic heron over a bamboo forest. Pure energy particles on wings and tail.

    Apenas 20% da qualidade real. Para 8K, DM.
    Only 20% of real quality. For 8K, DM me.

    #AIArt #FluxAI #Flux1 #LocalAI #4Steps #GTX1660Ti #ComfyUI #Atmospheric #CinematicLighting #HyperRealism #DarkFantasy #7Nodes #NoAPI #FOSS

  2. IA local é sobre controle e precisão. Produzo arte de alta fidelidade sem depender de nuvens ou filtros externos. Domínio técnico acima de tudo.

    Local AI is about control and precision. I deliver high-fidelity art without relying on cloud APIs or external filters. Technical mastery above all.

    #AIArt #FluxAI #Flux1 #LocalAI #4Steps #GTX1660Ti #ComfyUI #Atmospheric #CinematicLighting #HyperRealism #DarkFantasy #7Nodes #NoAPI #FOSS

  3. 🎉🎨 BREAKING: Black Forest Labs releases FLUX.1 Kontext dev, proving that even AI models deserve a midlife crisis by shedding their proprietary chains! 🤯🔓 Now you too can experience "proprietary-level" mediocrity from the comfort of your own consumer hardware—just don't make any money with it. 💸❌
    bfl.ai/announcements/flux-1-ko #BlackForestLabs #FLUX1 #AIrelease #OpenSource #ConsumerTech #MidlifeCrisis #HackerNews #ngated

  4. FLUX.1 Kontext: Neue Ära KI-basierter Bildbearbeitung made in Germany

    • Kombiniert Text & Bild-Input
    • Schnelle, präzise Bearbeitung
    • Hohe Konsistenz bei Figuren
    • Stilübertragungen in Sekunden
    • Beta mit Open-Weights verfügbar

    #ai #ki #artificialintelligence #blackforestlabs #flux1 #bildbearbeitung

    Jetzt LIKEN, teilen, LESEN und FOLGEN! Schreib uns in den Kommentaren!

    kinews24.de/flux-1-kontext-ki-

  5. Habe gerade mal meine #DrawThings App zu KI-basierten Bildgenerierung aktualisiert und lade mit das Modell #Flux1 schnell (8bit) herunter. Mal sehen, wie das im Vergleich zu #StableDiffusion lokal funktioniert.

  6. Advanced Image Editing Suite Enhances FLUX.1 Capabilities 🎨

    🎯 #BlackForestLabs launches #FLUX1 Tools suite with four key features: #Fill for state-of-the-art in/outpainting, #Depth for structural guidance, #Canny for edge-based control, and #Redux for image variations

    🔬 All tools available in two versions: #opensource [dev] models on #HuggingFace and professional variants via #API

    🏆 Benchmarks show FLUX.1 Tools outperforming competitors like #Ideogram and #Midjourney in respective categories

    🤝 Models accessible through partners including #falai, #Replicate, #TogetherAI, #Freepik, and #kreai

    💻 Complete documentation, weights, and inference code available on #GitHub

    blackforestlabs.ai/flux-1-tool

  7. Advanced Image Editing Suite Enhances FLUX.1 Capabilities 🎨

    🎯 #BlackForestLabs launches #FLUX1 Tools suite with four key features: #Fill for state-of-the-art in/outpainting, #Depth for structural guidance, #Canny for edge-based control, and #Redux for image variations

    🔬 All tools available in two versions: #opensource [dev] models on #HuggingFace and professional variants via #API

    🏆 Benchmarks show FLUX.1 Tools outperforming competitors like #Ideogram and #Midjourney in respective categories

    🤝 Models accessible through partners including #falai, #Replicate, #TogetherAI, #Freepik, and #kreai

    💻 Complete documentation, weights, and inference code available on #GitHub

    blackforestlabs.ai/flux-1-tool

  8. Advanced Image Editing Suite Enhances FLUX.1 Capabilities 🎨

    🎯 #BlackForestLabs launches #FLUX1 Tools suite with four key features: #Fill for state-of-the-art in/outpainting, #Depth for structural guidance, #Canny for edge-based control, and #Redux for image variations

    🔬 All tools available in two versions: #opensource [dev] models on #HuggingFace and professional variants via #API

    🏆 Benchmarks show FLUX.1 Tools outperforming competitors like #Ideogram and #Midjourney in respective categories

    🤝 Models accessible through partners including #falai, #Replicate, #TogetherAI, #Freepik, and #kreai

    💻 Complete documentation, weights, and inference code available on #GitHub

    blackforestlabs.ai/flux-1-tool

  9. Advanced Image Editing Suite Enhances FLUX.1 Capabilities 🎨

    🎯 #BlackForestLabs launches #FLUX1 Tools suite with four key features: #Fill for state-of-the-art in/outpainting, #Depth for structural guidance, #Canny for edge-based control, and #Redux for image variations

    🔬 All tools available in two versions: #opensource [dev] models on #HuggingFace and professional variants via #API

    🏆 Benchmarks show FLUX.1 Tools outperforming competitors like #Ideogram and #Midjourney in respective categories

    🤝 Models accessible through partners including #falai, #Replicate, #TogetherAI, #Freepik, and #kreai

    💻 Complete documentation, weights, and inference code available on #GitHub

    blackforestlabs.ai/flux-1-tool

  10. Advanced Image Editing Suite Enhances FLUX.1 Capabilities 🎨

    🎯 #BlackForestLabs launches #FLUX1 Tools suite with four key features: #Fill for state-of-the-art in/outpainting, #Depth for structural guidance, #Canny for edge-based control, and #Redux for image variations

    🔬 All tools available in two versions: #opensource [dev] models on #HuggingFace and professional variants via #API

    🏆 Benchmarks show FLUX.1 Tools outperforming competitors like #Ideogram and #Midjourney in respective categories

    🤝 Models accessible through partners including #falai, #Replicate, #TogetherAI, #Freepik, and #kreai

    💻 Complete documentation, weights, and inference code available on #GitHub

    blackforestlabs.ai/flux-1-tool

  11. Breaking News: Flux1.1 Pro Ultra auf Scenario

    Black Forest Labs (BFL) hat vor wenigen Tagen, am 6. November 2024, einen brandneuen Ultra-Modus für das Flux1.1 Pro AI-Bildmodell veröffentlicht. Dieser neue Modus kann bis zu viermal höhere Bildauflösungen als der Standardmodus erzeugen. Und seit heute ist dieses Modell auch auf Scenario verfügbar. kantel.github.io/posts/2024111 #Scenario #Flux1 #Bildgenerator #InteractiveFiction #GameDev

  12. Noch mehr OpenArt.ai-Tests (diesmal als Comic)

    Weil ich Blut geleckt hatte und weil mir die im letzten Beitrag verwendete LoRA Vibrant Fantasy einmal zu realistische Bilder erzeugte und zum anderen Probleme mit den Fingern hatte, habe ich noch ein paar Credits verballert und eine Handvoll weiterer Bilder der nackten Kriegerin, die gegen einen Alien kämpft, von OpenArt.ai erzeugen lassen. kantel.github.io/posts/2024110 #OpenArt #Flux1 #InteractiveFiction #GameDev

  13. New KIs on the Blog: OpenArt.ai

    OpenArt.ai ist ein relativ neuer, leistungsfähiger Bildgenerator, der in der Hauptsache auf die KI-Modelle Stable Diffusion XL und Flux.1 [dev] setzt. Der Dienst ist nach einer einmaligen, 40 Credits bereitstellenden Startphase kostenpflichtig. kantel.github.io/posts/2024110 #OpenArt #Bildgenerator #StableDiffusion #Flux1 #InteractiveFiction #GameDev

  14. Zurück ins Wunderland: Little Girl Alice

    Inspiriert durch das Flux-Fieber, das mich gepackt hatte, habe ich mich gestern abend hingesetzt, und in Scenario meine erste Flux-LoRA trainiert. Ziel war es, aus dem Modell »Expressive Comic Characters« ein Character Model zu generieren, das möglichst konsistente Bilder eines kleinen Mädchens erzeugt, das von Lewis Carrolls »Alice im Wunderland« inspiriert ist. kantel.github.io/posts/2024102 #Scenario #Bildgenerator #Flux1 #InteractiveFiction #Twine #GameDev

  15. Flux und Scenario: Ein echter Game Changer?

    Flux.1, der angeblich »neue Stern am Himmel der bildgenerierenden gekünstelten Intelligenzia«, war ja schon zwei mal Thema in diesem Blog Kritzelheft. Und auch Scenario, die derzeitige Künstliche Intelligenz meines Vertrauens, hat – erst vorsichtig, dann mit wachsender Begeisterung – Flux.1 in ihr Portfolio integriert. kantel.github.io/posts/2024102 #Scenario #Flux1 #InteractiveFiction #GameDev #Bildgeneratoren

  16. If you give #FLUX1.1 a prompt like "IMG_1018.CR2" you get back images that are so very hard to tell they're AI.

    The realism here is kicked up a notch.

    x.com/fofrai/status/1841854401

  17. Konsistente Charactere mit Scenario

    Wer meinem Stream auf Flickr folgt, hat sicher mitbekommen, daß ich momentan sehr viel Zeit damit zubringe, Scenario, der derzeitigen gekünstelten Intelligenzia meines Vertrauens, beizubringen, konsistente Charaktere zu erzeugen, die in Spielen und interaktiven Geschichten verwendet werden könnten. kantel.github.io/posts/2024100 #Bildgenerator #KünstlicheIntelligenz #InteractiveFiction #GameDev #Scenario #StableDiffusion #Flux1

  18. Milch macht müde Männer munter? Flux.1 bei Scenario

    Flux.1 ist ja so etwas wie ein neuer Stern am Himmel der bildgenerierenden gekünstelten Intelligenzia. Die Software soll nicht nur mit den Boliden MidJourney, Googles Imagegen oder DALL-E 3 mithalten können, sondern sie ist in der Grundversion Flux.1[schnell] auch noch Open Source (Apache 2.0 Lizenz). kantel.github.io/posts/2024092 #Bildgenerator #KünstlicheIntelligenz #InteractiveFiction #GameDev #Scenario #Flux1

  19. #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

  20. 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 model. Did not disappoint...

    1/7

  21. #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

  22. #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