Pikaeffects lets you inflate, crush, squish, melt, and explode any object in the video.
2024 is getting owned by AI video generators!
In just the past few months, we’ve seen a whirlwind of activity in the AI video space. Runway released a new video-to-video model, Kling AI rolled out Kling 1.5 with motion-brush and lip-syncing features, and Vidu launched new video models with super-fast generation speed.
Today, Pika joins the club with the release of its new video model, Pika 1.5.
Pika is a generative AI platform that allows you to create videos from text or image input. The tool isn’t new to the AI video scene. In fact, I’ve been keeping an eye on it and have covered their model releases in previous posts.
Back in December, when Pika launched their 1.0 model, I noted how groundbreaking it was for its time. If you’re curious, here’s an article I wrote about Pika 1.0.
Pika was co-founded by Demi Guo and Chenlin Meng, former PhD students at Stanford University’s Artificial Intelligence Lab. Their vision was clear from the start: to revolutionize video creation through AI. And they’ve delivered on that promise in more ways than one.
Recently, they raised an $80 million Series B round led by Spark Capital, bringing their total funding to $135 million. This funding will accelerate their efforts in building the best AI video foundation model, positioning them to become a key player in the creative technology sector.
Demi Guo, one of the co-founders, had this to say about their journey so far:
“When we started Pika six months ago, we wanted to push the boundaries of technology and design a future interface of video making that is effortless and accessible to everyone. Since then, we’re proud to have grown the Pika community to half a million users, who are generating millions of videos per week.”
Upon login, the first thing I noticed is the revamped user interface. It’s much cleaner, more intuitive, and noticeably faster than its predecessor.
Here’s a breakdown of the major updates in Pika 1.5:
There are two primary ways to access Pika:
On Discord, you can generate videos for free by going into one of the “creations” channels and typing /create
to toggle the prompt field.
You can describe the video you want to create, and within a few minutes, the AI processes and delivers your video.
If you want to animate an existing image, use the /animate
command to add motion to your input image.
For an easier and more streamlined experience, you can use Pika directly on their website. The interface here is modern and easy to navigate. Upon signing in, you’ll see a modal window announcing the release of Pika 1.5. From there, it’s as simple as typing your prompts into the field and letting the AI work its magic.
You have the prompt field on the bottom section and a tab for your generated videos and community-generated videos in the center.
Here’s an example:
Prompt: dolly in to a chameleon is doing push ups
Pretty cool, right?
You can also make adjustments on the negative prompts (specify what you don’t want to see on the final result), seed (a number as a reference for the output), and the aspect ratio of your video.
Right now, only the paid subscribers are allowed to switch back to the older model, Pika 1.0. All users are defaulting to Pika 1.5.
Now, let’s talk about Pikaffects. This is easily one of the most fun aspects of the new release, and it’s where Pika stands out from its competitors.
Pikaffects makes it super easy and fun to get the most out of our model without any complex prompts. You can inflate, crush, squish, melt… do whatever you like!
To start using this tool, click on the Pikaeffect button, and a set of effects panel will be displayed. From there, you can inflate, crush, squash, or melt any object in the video with just a click.
Here’s an example with the cake slice effect:
Great. I’d love to show you more examples, but the videos are already making the size of the article too large. I encourage you to try them out for free on Pika Lab’s website. I’m sure you’re going to have a lot of fun.
On signup, you’ll get 250 free credits. Each 5-second video generation costs 15 video credits. Get 30 new credits every day after using your initial 250.
If you run out of credits, you can opt to upgrade to a paid subscription. Pika’s pricing is pretty straightforward, and there’s a plan for just about everyone.
Now, keep in mind that video credits don’t roll over month-to-month, which is a bit of a downside if you don’t use them all. But if you opt for the annual plan, you get a 20% discount, which is a nice bonus.
Overall, I am very impressed with how Pika 1.5 handles photorealism, dynamic physics, and sophisticated camera techniques. Small businesses and content creators can create high-caliber ads on a budget, making Pika 1.5 an affordable way to produce visually stunning promotional videos.
The Pikaeffects feature may seem like gimmicks at first glance, but in reality, they provide a lot of creative freedom to users who might not have the technical know-how to craft such videos manually.
When it comes to the pricing, the Standard plan offers the best bang for your buck if you’re creating on a moderate level, but if you’re serious about content creation, the Pro or Unlimited plans are where you’ll really see the value, especially with the flexibility and speed you get.
That’s about it. What do you think about Pika 1.5? If you’ve been keeping up with the recent AI video announcements, how does Pika stack up with Runway and Kling? I’d love to know your thoughts.
PS. I am planning to do a comparison between Pika, Runway, Kling, and Vidu in future articles. If you think this is interesting, please let me know in the comments so I can prioritize it if there’s enough interest.
‍
Software engineer, writer, solopreneur