OpenAI’s highly anticipated video generation model Sora officially launched today. As long as you have a ChatGPT Plus or ChatGPT Pro plan, you can use Sora to generate videos without any additional subscription or payment. However, since Sora just launched and everyone wants to try it out right away, the registration volume is so high that OpenAI has temporarily paused new Sora registrations. Although you can’t sign up right now, we can still check out videos that other lucky users have generated with Sora, so everyone can enjoy what others have created before trying it out themselves. I’ve rounded up some impressive videos today—if you’re interested, keep reading~
Sora is disgustingly good, gj @sama
Adding it to @everartai asap so you can bring your images to life pic.twitter.com/wMehxOc8cm
— Jay A (@jay_azhang) December 9, 2024
OpenAI’s text-to-video tool Sora is officially out! It’s so popular I can’t even get in line for it, so let’s check out what netizens have made first.
Today OpenAI Officially launching its own video generation model, SoraPerhaps because it had been promoted for so long and can now finally be used, many people can’t wait to try it out. On top of that, there’s no need for a separate subscription or payment—just being a ChatGPT paid user is enough. This led to all ChatGPT paid users rushing to register for a Sora account, which is also why OpenAI has now paused accepting new user registrations for Sora. Anyone who wants to sign up has to wait in line for their turn. Some people still can’t get in even after Sora launched this morning. For those who still can’t access it or haven’t purchased a ChatGPT paid plan, today the editor has put together some videos generated by netizens using Sora, so everyone can enjoy others’ work while waiting in line to get in.
- Sora can present the effect of switching back and forth between different shots within a single video, and the characters maintain consistency, with no frame drops:
Sora をおためし中https://t.co/LKFuZTmz3T
cute cat ears maid of japanese anime style pic.twitter.com/atav15TMRN— 布留川英一 / Hidekazu Furukawa (@npaka123) December 9, 2024
- Sora can generate very realistic news broadcast footage, including editing, moving captions, news-style camera shots, anchor lip movements, and other details. However, it still cannot overcome the problem of garbled characters appearing in titles within the videos.
The rumors are true – SORA, OpenAI’s AI video generator, is launching for the public today…
I’ve been using it for about a week now, and have reviewed it: https://t.co/jII49vkuHN
THE BELOW VIDEO IS 100% AI GENERATED
I’ve learned a lot testing this, here are some new… pic.twitter.com/uA1EhRuK7B
— Marques Brownlee (@MKBHD) December 9, 2024
- Sora handles animal details quite well—the monkey’s eyeballs, fine lashes, and blinking are so detailed that if you weren’t told, you might really think it was live-action footage shot by humans.
3. Soulful monkey serenity pic.twitter.com/u40j1A2gUz
— Min Choi (@minchoi) December 9, 2024
- Sora can also generate gameplay footage. Although the details still need improvement, the results are quite impressive.
sora使えるようになってた！
ひとまず勇者とドラゴンの戦闘動画を生成してもらった！かなり良い！！ pic.twitter.com/gsf25jgHGD— gimu@AI×ゲーム開発 (@gimu_ai) December 9, 2024
- Sora seems to have an option called “Blend” that can merge separate scene and character action shots together:
Sora’s “Blend” option might be my favorite feature so far. I wanted to see how it would handle two very different shots. pic.twitter.com/eqj5pPxPbb
— Blaine Brown (@blizaine) December 9, 2024
- The well-known Japanese AI creator @KakuDrop also shared videos generated with Sora. As expected of a master, the resulting characters already have a very distinctive style.
〜 Echoes of Grace 〜 Full version 🔈🔈🔈
🎨 #sora
以前ポストした作品のフルバージョンをお届けします🚀
@OpenAI #genai #動画生成ai pic.twitter.com/gGxIrdcHFg
— Kaku Drop 架空飴 (@KakuDrop) December 10, 2024
- Sora does a pretty good job with the backgrounds—like the streets of 80s Harajuku in Japan, where the clothing worn by the people in the video really has the flavor of that era. But if you look closely, there are still quite a few flaws, such as a background woman who looks like a giant, or a camera shot that passes right through between two people’s heads.
This is wild.
OpenAI just dropped Sora, their state-of-the-art AI video model.
The video quality looks insane.
100% AI, these are not real.
1. 80s Japan Harajuku fashion street pic.twitter.com/Q8xXIbdP4z
— Min Choi (@minchoi) December 9, 2024
I wonder how everyone feels about Sora’s performance after watching the video shared by the netizen above? Sora has been hyped for 10 months since its announcement. Compared to other video generation models, Sora really can be considered one that took a long time to launch after promotion. Presumably, the longer the promotion, the higher everyone’s expectations for Sora. Although there are too many people wanting to use Sora right now, so we probably won’t get a turn today, but maybe we can use it tomorrow. Friends interested in Sora can queue up first, so they can immediately experience the effect of generating videos with Sora when the time comes.
Source: KOCPC Chinese