EasyGen a republié ceci
I test OpenAI's '𝗦𝗼𝗿𝗮' vs. the new Google '𝗩𝗲𝗼-𝟮'. I feel like comparing a 𝗯𝗶𝗸𝗲 vs. a 𝘀𝘁𝗮𝗿𝘀𝗵𝗶𝗽: → 𝗪𝗵𝗮𝘁 𝘆𝗼𝘂 𝗰𝗮𝗻 𝘀𝗲𝗲 𝗼𝗻 𝘁𝗵𝗲 𝗿𝗶𝗴𝗵𝘁 I generated videos, just from text, using Sora. It's the most hyped AI video generator from OpenAI. And the results... are dicey. → 𝗪𝗵𝗮𝘁 𝘆𝗼𝘂 𝗰𝗮𝗻 𝘀𝗲𝗲 𝗼𝗻 𝘁𝗵𝗲 𝗹𝗲𝗳𝘁 This is the new model from Google: Veo-2. It's not live yet, but I extracted tons of examples. As you can see, the results are fascinating: ☑️ Ultra-realistic physics ☑️ Complex camera motion ☑️ Multiple characters with vibrant emotions → 𝗪𝗵𝗮𝘁'𝘀 𝗺𝘆 𝗰𝗼𝗻𝗰𝗹𝘂𝘀𝗶𝗼𝗻? 2025 is the year of AI video becoming a thing. It's no longer a meme. It's no longer funny. It expands creativity just like ChatGPT did to writing. ♻️ Repost this video to inform your network.