Vinija Posted June 17 Share Posted June 17 Luma AI released their new text to video model available here:Â https://lumalabs.ai/dream-machine/ I'm attaching a few I've generated, give it a whirl and share your creations! watermarked_video063fc7327f81c47f6ae02927d82c5c5b3.mp4 watermarked_video033876e77ed084be08b1e40c40aa7f01a.mp4 1 Quote Link to comment Share on other sites More sharing options...
subin_vidhu Posted June 17 Share Posted June 17 Hey @Vinija, May I know the time it took to generate these videos? I have tried it a few days back and it took quite some time for the results. Quote Link to comment Share on other sites More sharing options...
subin_vidhu Posted June 17 Share Posted June 17 Here are some of the outcomes I observed when I put it to the test: Prompt : AI in medical domain https://storage.cdn-luma.com/lit_lite_inference_text2vid_v1.0/0646e926-6f59-4446-b9b7-1392a6e04b12/watermarked_video00e7c0416cf1e498bbf1e26830e547863.mp4 Â simple.mp4 Â Â Prompt:Â Envision a futuristic medical facility in the year 2035, where AI has revolutionized healthcare. Describe the scene, highlighting the integration of AI-powered technologies in medical imaging, diagnostics, and patient care. Show me the futuristic medical equipment, augmented reality interfaces, and collaborative robots working alongside human healthcare professionals to provide personalized and precise care. https://storage.cdn-luma.com/lit_lite_inference_text2vid_v1.0/eba4c2eb-34c4-47ac-bacf-bc6d55e0af4e/watermarked_video03c4ae61810d14d6aae0e4e72abd5dbf5.mp4 Â adv.mp4 Â Quote Link to comment Share on other sites More sharing options...
Vinija Posted June 17 Author Share Posted June 17 Yes, it too, quite a long time. It was queued first then generated so do you need patience for this one. Yours are pretty cool too but I've noticed the model glitches with human details like fingers often. Quote Link to comment Share on other sites More sharing options...
subin_vidhu Posted June 18 Share Posted June 18 Hopefully these glitches may get corrected in the coming models, I came across this post https://runwayml.com/blog/introducing-gen-3-alpha/ today and it seems that the runway team have come up with up a Gen 3 model for video generation. I don't know if it is public yet, but I tried out their Gen 2 model and for the second prompt I used for LUMA, the output was: runway.mp4 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.