I need it to actually generate unique situations (eg me in a car/me on a plane) so it needs to be a diffusion transformer type approach
MULTIMODAL AI
-
Testing Flux LoRA Animation Techniques for Dynamic Video Generation
By
–
I have a hunch and it requires an experiment to test it out. For dynamic shots, we sometimes use flux Lora (trained on my face) + Runway to animate + Lipsync AI to sync audio However single photo Loras don’t animate well especially when angles change. We want to instead try to
-
Genie 2: Large-Scale Foundation World Model for AI
By
–
Genie 2: A large-scale foundation world model https://
bit.ly/40f6QHv
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
xAI Launches Standalone Grok iOS App with Image Generation
By
–
ITS HERE: xAI’s brand-new standalone Grok iOS app.
— xAI (@xai) 10 janvier 2025
Harness powerful AI, generate stunning images, and login with X to personalize your experience with real time news, sports and local data.
Download now in the US:https://t.co/5Ejdqh8a34 pic.twitter.com/d8jnR4ZjmDITS HERE: xAI’s brand-new standalone Grok iOS app. Harness powerful AI, generate stunning images, and login with X to personalize your experience with real time news, sports and local data. Download now in the US: https://
apps.apple.com/us/app/grok/id
6670324846
… -
OpenAI developing video search for Sora
By
–
WIP : OpenAI is working on a video search for Sora There you will be able to search for similar videos from the recent feed. This feature is not yet available to all users.
-
North Leverages Compass Multimodal AI Search System
By
–
North also leverages Compass, our most advanced multimodal AI search and discovery system to help employees quickly and accurately surface relevant insights from across their business.
-
Krea Launches Video Models and Video-to-Audio Generation
By
–
introducing new video models and video-to-audio.
— KREA AI (@krea_ai) 9 janvier 2025
Kling 1.6, Hunyuan, and 01-Live are now available in Krea and now you can generate audio for videos.
free users can finally generate videos too! pic.twitter.com/CgB9f2hk8Tintroducing new video models and video-to-audio. Kling 1.6, Hunyuan, and 01-Live are now available in Krea and now you can generate audio for videos. free users can finally generate videos too!
-
ChatLLM Teams: Interact with Multiple Data Modalities Using AI
By
–
Interact with many data modalities (PDF/text, images, dashboards,…) using @AbacusAI ChatLLM Teams. This video highlights DataLLM, which allows you to chat with (ask questions to) your data! Sign up for FREE TRIAL here: https://t.co/ATIFqtJ0Di
— Kirk Borne (@KirkDBorne) 8 janvier 2025
———#LLMs #GenAI #AI #GenerativeAI pic.twitter.com/XdwXFaigyUInteract with many data modalities (PDF/text, images, dashboards,…) using @AbacusAI ChatLLM Teams. This video highlights DataLLM, which allows you to chat with (ask questions to) your data! Sign up for FREE TRIAL here: http://
chatllm.abacus.ai/?token=kirk
———
#LLMs #GenAI #AI #GenerativeAI -
Google Launches PaliGemma 2 Vision Language Models
By
–
Welcome PaliGemma 2 – New vision language models by Google https://
bit.ly/3VQsWNT
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Open Source Vision Language Models: SmolVLM, MOLMO, Qwen VL
By
–
You should use a VLM. SmolVLM, MOLMO or Qwen VL are good open source starting points little bot