Pika 1.5 launches with physics-defying AI special effects

Share This Post


Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More


Pika, also known as Pika Labs, was one of the first startups to emerge with its own AI video generation model, allowing users to simply type in text and get video clips in return, raising $35 million in a Series A in November 2023.

But it’s been nearly a year since Pika launched its 1.0 text-to-video AI platform and since that time, many competitors have emerged and/or updated their models to outclass Pika in terms of realism and the effects they’re capable of, chief among them, Runway which is on its Gen-3 Alpha Turbo model and Luma AI which in August debuted its Dream Machine 1.5 AI video model.

But no longer: acknowledging its relatively lengthy (at least for the AI industry) period of quiet, Pika today announced it is launching Pika 1.5, an updated version of its model that offers eye-popping, physics defying special effects or “Pikaffects” that can transform imagery subjects into bizarrely malleable versions of themselves.

These animated special effects are available from a new button labeled Pikaeffects, and include the following pre-sets: Explode it, melt it, crush it, inflate it and “cake-ify it” — the latter of which essentially turns your image subject into the meme videos of highly realistic prop cakes.

Pika 1.5 will automatically seek to identify the subjects or objects in the video and apply the corresponding effect, even if it’s not possible for that subject or object to transform in that way in reality.

Some of the effects — namely crush it, squish it, and cake-ify — actually insert new props such as a hydraulic press, human hands and a knife into the frame, yet allow them to interact with the objects in the still image and thus, resulting video.

The results are often hilariously incongruous yet convincing, with Pikaffects turning the entire world into a playground of convincing yet surprising deformations.

Already winning fans among AI video creators and early adopters

Already, AI early adopters and video creators on X are posting incredible results and noting that while other rival AI video generators have largely all raced to provide increasing realism and more control of the virtual “camera” in their platforms, Pika is laudable for pursuing a radically different approach.

Starting today, both free and paid users can access this new version, unlocking an array of advanced features and creative possibilities.

More improvements across the board

Pika 1.5 promises significant improvements across the board, particularly in generating more powerful video clips through both image-to-video (i2v) and text-to-video (t2v) workflows.

Users can now create high-quality, five-second clips, incorporating lifelike movements such as running, skateboarding, and even flying.

For those looking to make their creations truly cinematic, the platform has added new motion control features, allowing users to include advanced shots such as Bullet Time, Vertigo, Dolly Left, and Crane Down.

This makes it easier to capture dynamic, professional-level footage without needing deep technical expertise.

Paid users can still use Pika 1.0 and its Lip Sync and AI Sound Effects

While Pika 1.5 offers a range of exciting new features, paid users still have the option to switch between the older Pika 1.0 model and the new version.

Features such as its ElevenLabs-powered Lip Sync, SoundFX, and the ability to extend and expand clips remain tied to Pika 1.0 for now.

The platform has also made it easy to control camera motion, styles, and effects directly within text prompts, offering more creative flexibility.

Pricing remains unchanged — but the number of credits to make a video is going up

The company has confirmed that while subscription prices remain unchanged, each five-second clip generation will now require 15 credits, reflecting the increased power and resource demands of the new model.

Users will notice that generation times are naturally longer for the Pika 1.5 model, though Pikaffects is designed to generate faster, offering a quicker way to explore the new capabilities.

Pika is actively encouraging users to experiment with Pika 1.5 by joining its community challenges, where participants can win free credits to use on the platform. Credits won through these challenges are fully valid for use with the latest model. Additionally, users are invited to share their feedback and report bugs via the platform’s feedback channels, helping Pika continue to improve its tools.

With the release of Pika 1.5, users now have access to a more powerful model for creating high-quality, dynamic video content.

The addition of Pikaffects, cinematic shots, and lifelike movements offers exciting new possibilities for creators, regardless of their experience level, and is clearly already helping Pika differentiate itself amid an increasingly crowded and competitive AI video model space.



Source link
spot_img

Related Posts

X changes revenue model to reward paying users, moving away from ads

Musk, despite promising earlier to remain politically neutral...

Surviving Crypto Ups and Downs: Insiders Share Mental Health Guide for Investors

As global regulatory measures shape the evolving crypto...

Bitcoin falls below $61,100 ahead of US inflation data; Solana, Shiba Inu drop up to 4%

Cryptocurrency prices declined on Thursday as markets awaited...

Unisys expands in India, investing in AI and quantum computing

Even amid the disruptions in global supply chains...

NASA “really looking forward” to next Starship test flight

Updated 9 p.m. Eastern with Gerstenmaier comments. WASHINGTON —...

New technique makes RAG systems much better at retrieving the right documents

By adding knowledge of surrounding documents to document...
spot_img