i still look at other stuff, but i care a lot about the "why now", so end up leaning heavily in AI today
GENERATIVE AI
-
Runway ML Veo3 Identity Cold Start Problem Solutions
By
–
yeah i used @runwayml but veo3 under the hood identity is an interesting space today, but cold start problem is still hard. would be curious how you're solving for this (just followed so DM me)
-
DeepSeek-V3.1-Terminus Released on Hugging Face with Improvements
By
–
DeepSeek-V3.1-Terminus is out on Hugging Face
— AK (@_akhaliq) 22 septembre 2025
This update maintains the model's original capabilities while addressing issues reported by users, including:
Language consistency: Reducing instances of mixed Chinese-English text and occasional abnormal characters
Agent… pic.twitter.com/bw9AtNHFUrDeepSeek-V3.1-Terminus is out on Hugging Face This update maintains the model's original capabilities while addressing issues reported by users, including: Language consistency: Reducing instances of mixed Chinese-English text and occasional abnormal characters Agent
-
Creating Video with Runway AI: Cost, Time, and Voice Consistency
By
–
cost about ~$120 in Runway credits, finished 95% of it in an afternoon (3-4 hours while doing other stuff) voice consistency of character throughout video was a challenge so i gave up overall, pretty happy with it!
-
ChatGPT and Runway Generate Video Scripts Automatically
By
–
and then for each scene, i used chatgpt to generate the two characters in a different scene, and prompted runway to generate the script. one sentence per character per generation worked pretty well
-
Comparing AI Video Tools for Multi-Person Dialogue Generation
By
–
i then tried a couple of tools to generate the dialogue, but couldn't quite get there with hedra (great for one person, but not two+), multitalk (this model is specifically for this, but i wasn't getting the quality i wanted), grok, sora, gemini, and ended up settling on runway -… pic.twitter.com/LmmOiLVps0
— Yohei (@yoheinakajima) 22 septembre 2025i then tried a couple of tools to generate the dialogue, but couldn't quite get there with hedra (great for one person, but not two+), multitalk (this model is specifically for this, but i wasn't getting the quality i wanted), grok, sora, gemini, and ended up settling on runway –
-
Controlling AI dialogue consistency and speaker attribution challenges
By
–
even on runway, if the dialogue went back and forth too much, it didn't keep consistency on who speaks, and was hard to control
— Yohei (@yoheinakajima) 22 septembre 2025
(example video here messes this up) pic.twitter.com/IMt6vgpyhReven on runway, if the dialogue went back and forth too much, it didn't keep consistency on who speaks, and was hard to control (example video here messes this up)
-
Artist Generates Surreal Car Scene Using AI Image Generation
By
–
this was just an idea at first, i started by generating an image of a unicorn and alien in the back of a car
-
Creating AI-Generated Video Content for $120 Using Runway ML
By
–
for anyone curious about the video…
— Yohei (@yoheinakajima) 22 septembre 2025
took about ~$120 using @runwayml to generate the clips, images by chatgpt, edited in canva
full process 👇 https://t.co/sxxAsnwOkLfor anyone curious about the video… took about ~$120 using @runwayml to generate the clips, images by chatgpt, edited in canva full process
-
AI Auto-Drafts Responses Matching Your Tone Style
By
–
It also auto-drafts responses that match your tone and style. pic.twitter.com/BeBK9UuOpq
— Perplexity (@perplexity_ai) 22 septembre 2025It also auto-drafts responses that match your tone and style.