“I owe you my life, ser.” Loras Tyrell “I am no ser.” Sandor Clegane

LTX team doesnt stop shipping 🤯 All new LTX 2.3 IC LoRAs just dropped on HuggingFace🔥

Use better systems! Best comfyui, there you have all, open source and commercual models. Flux klein for example, open source , in special with loras, does a lot better jiob than grok. Often also than nb2 or pro. I am sure u known all about comfyui, if not dm me,

'Significant momentum:' Loras wraps up $20 million cash campaign telegraphherald.com/news/tri…

Look Renly we love you man but You adn Loras is 20 and 15 old. I dont think you cant clean one sin to try clean with sin.

Welcome to my page. I create AI illustrations focused on anime aesthetics and fashion design. I train LoRAs to develop new styles and explore different visual expressions. New artwork is posted regularly. Thank you for your continued support. Thank you for the follows, likes, and reposts. A special thank you to all my followers in Japan. Feel free to repost my work. Thank you for visiting. ようこそ私のページへ。 アニメ風の表現とファッションデザインを中心に、AIイラストを制作しています。LoRAの学習を通して、新しいスタイルやさまざまなビジュアル表現を探求しています。 作品は定期的に投稿しています。いつも応援していただき、ありがとうございます。 フォロー、いいね、リポストしてくださる皆さまに感謝いたします。 日本のフォロワーの皆さまにも、心より感謝申し上げます。 作品のリポストはご自由にどうぞ。 ご覧いただきありがとうございます。

Most AI video tools stop at prompting, but LTX goes a layer deeper. Introducing the new LTX Trainer. Train LoRAs and IC-LoRAs across video, audio, cross-modal, and reference-conditioned workflows from a single framework. Plus: • New agentic skill • Flexible conditioning • Free IC-LoRAs • Fully open source Most models are generic. The best ones become yours. GitHub: github.com/Lightricks/LTX-2/… Documentation: docs.ltx.video/open-source-m… HuggingFace: huggingface.co/collections/L…

Ran into an issue with Anima lately: using character LoRAs forces single-character compositions. I troubleshot this by making my own LoRA and found the fix: your training dataset can't just be solo shots. You MUST include images of the main character in the same frame with others, and use natural language for tagging! The big plus? It saves way more tokens than the old Illustrious LoRAs. We used to need 20 repeats & 10 epochs, but now a maximum of 10 repeats gives amazing results! Really hoping TensorArt updates their defaults soon. Using the old system's settings on the new system is just terrible right now. #Marriagetoxin #KillOrKiss #HikaruGero #AIart

Las loras de la derecha👇 Pasas desapercibida. Nadie te llorará.

I have been training loras for ideogram 4 so no time for new models. Ideogram said they are releasing edit model at somepoint and that would be cool

Bouaddi lo has visto 45 minutos y están como loras repitiendo su nombre

Lucía Loras, 𝐑𝐞𝐛𝐞𝐥 𝐟𝐢𝐧𝐬 𝐚𝐥 𝟐𝟎𝟐𝟕 👏

[buenos días gente bonita, aquí más de está Vtuber que no quería r34 de su avatar xD, además, esta pose es de otro Lora que hice, hice 4 loras, el de anoche con las 2 chicas, este y otros dos mas] {deme -- vtuber}

my favorite of the new LTX 2.3 LoRAs ✨ huggingface.co/Lightricks/LT… Remember when a year ago we need a full on model like VACE to do that? and now an IC LoRA is even better