A startup called Pika Labs, which is only six months old and works on video AI, shared that it successfully gathered $35 million in funding during its series A round. Lightspeed Venture Partners led this funding round, and combined with previous rounds, Pika Labs has now secured a total of $55 million.
In addition to the funding news, the company introduced Pika 1.0, a new web platform. This platform builds upon an earlier beta release and allows users to create and edit videos in various styles like 3D animation, anime, or cinematic all based on text prompts. You can check out some examples in the promotional video published by Pika, which we have shared below.
This update is the newest in the quickly changing world of AI-powered video creation. Pika is now in direct competition with similar tools from Runway and Stability AI. Adobe is also exploring this field and has presented several generative AI tools capable of generating and editing objects within videos.
Pika has shared that its product, already in use by over half a million people, is open for sign-ups. It will be released gradually, allowing users to become creative video directors and producers.
Pika’s rapid growth is reflective of the continued, strong demand for generative AI of all flavors from tools like Midjourney and DALL-E 3 to ChatGPT.
What to Expect From Pika 1.0?
Demi Guo and Chenlin Meng, who were once Ph.D. students in Stanford’s Artificial Intelligence Lab, introduced the initial version of Pika on the messaging app Discord in late 2023.
Since that time, the product has advanced into a sophisticated tool, producing millions of videos every week. The company reports that its top users on Discord spend up to 10 hours a day using its features to create videos.
This high level of user engagement has attracted investors, leading to the company securing $20 million in funding through pre-seed and seed rounds. Nat Friedman and Daniel Gross were among the leaders in these funding rounds.
Additional investors in the initial funding rounds included Elad Gil, Adam D’Angelo, Andrej Karpathy, Clem Delangue, Craig Kallman, and Alex Chung.
The CEO of the company, Guo, shared in a statement, “We understand firsthand the challenges of creating high-quality content, which can be both difficult and costly. We developed Pika to provide everyone, from everyday users to professional filmmakers, with the tools to bring high-quality videos to life. Our vision is to empower anyone to become the director of their stories and unleash the creator in all of us.”
Pika Labs has introduced Pika 1.0, the latest version of its video AI platform featuring a new AI model. This version provides users with a user-friendly web interface instead of relying on the Discord app. Through this interface, users can easily create and edit higher-quality videos using straightforward text prompts.
To generate the desired output, users only need to input the envisioned style, and the underlying model will produce the result within seconds. The tool is capable of creating various content, including 3D animations, live-action videos, cinematic clips, and can even modify moving objects like a horse or an outfit with simple text prompts.
From Text-to-Video to Image-to-Video
In addition to text-to-video, Pika 1.0 will now offer support for image-to-video and video-to-video capabilities. This means users can convert their existing content into various video styles.
For example, it could animate a meme or transform a cinematic video into an animated format. Users have the option to upload their content and provide text prompts for edits such as expanding the canvas or adjusting the aspect ratio of the clip. They can also guide the model to predict content beyond the original video, including adding new objects, props, and characters to create an entirely new video creation experience.
How to Access Pika 1.0?
According to a blog post by Pika Labs, the updated platform will be accessible on Discord and through a dedicated website for both mobile and desktop users. Interested individuals can visit the website and sign up, but it’s essential to be aware that the new video AI features won’t be immediately accessible. The company has implemented a waitlist and plans to gradually grant access to users who have signed up.
This approach is in place to avoid overloading the system.
Growing Competition in Video AI Space
With this funding, Pika Labs has reached a total capital of $55 million and a valuation of nearly $200 million. This firmly establishes the company as a significant player in the growing field of video AI, which has been gaining attention in recent months. Stability AI recently introduced Stable Video Diffusion models capable of producing high-quality clips from images, but it is limited by the length of the generated videos.
Pika’s AI tool, on the other hand, can create longer clips and even allows users to extend the length of the videos they upload. However, the quality of the extended videos is yet to be fully evaluated.
In just six months, Pika Labs, a startup focused on video AI, raised an impressive $35 million in funding, bringing its total funds to $55 million with a valuation close to $200 million. The upgraded Pika 1.0 platform now allows users to create diverse videos using text, images, and existing video content.