OpenAI’s SORA is the biggest talk of the town and with that, people are looking for its alternatives as well! Higgsfield is here to take the spot with its new Diffuse iOS application.
Highlights:
- Diffuse is a mobile application by Higgsfield AI where users can create videos using text prompts.
- Users can also create videos with their avatars by uploading their selfies.
- The iOS App is currently available in Diffuse is available in India, South Africa, the Philippines, and Canada.
Meet Higgsfield AI’s Diffuse App
Higgsfield AI is a Video AI company that wants to democratize social video creation for everyone. They recently released their first app called Diffuse on iOS, based on their foundational model that excels at creating personalized characters with lifelike motion.
Hi! We’re Higgsfield – a Video AI company that's democratizing social video creation to everyone.
— Higgsfield AI (@higgsfield_ai) April 3, 2024
Our game changing foundational model excels at creating personalized characters with lifelike motion – with just 1 selfie and all on mobile.
We bring any story to life. Watch👇 pic.twitter.com/b4BogCjwUF
One differential factor for Higgsfield is that it is a Transformer model, while other Text to Video AI technologies are based on Diffusion model.
The model offers a wide range of control which is all directed through text. Let’s look at some examples:
Prompt: A red fish swims through a vibrant coral reef. A group of colourful fish are swimming ahead.
Output:
Here's the full clip of the fish swimming through the coral reef found in the demo reel.
— Higgsfield AI (@higgsfield_ai) April 3, 2024
Prompt: A red fish swims through a vibrant coral reef. A group of colorful fish are swimming ahead. pic.twitter.com/GtIZ2MqSjE
Prompt: A colourful iguana sits still while looking ahead. Beams of bright light move across the surface.
Output:
Another 7 second clip generated by our model:
— Higgsfield AI (@higgsfield_ai) April 3, 2024
Prompt: A colorful iguana sits still while looking ahead. Beams of bright light move across the surface. pic.twitter.com/UTizBIPkun
Prompt: A man in a canoe floats down a calm river surrounded by huts and boats. The sun glistens on the water and peeks through mountains.
Output:
Our model offers a wide range of control, all powered by text.
— Higgsfield AI (@higgsfield_ai) April 4, 2024
Prompt: A man in a canoe floats down a calm river surrounded by huts and boats. The sun glistens on the water and peeks through mountains. pic.twitter.com/Wx5Lfv5XC4
Diffuse is available in India, South Africa, the Philippines, Canada, and Central Asia, but soon be rolled out to more countries worldwide.
Users can quickly create a video by selecting from a library of presets and a selfie. For original ideas, Diffuse has a built-in Prompt Builder that can take in a combination of text prompts, reference images or videos. SORA also can create videos from images, but we don’t know whether it will work with user’s photos or not.
The biggest advantage of Diffuse is that it is a mobile application. This removes the obstacle of having a deep level of technical knowledge and allows for our creative tools to get into the hands of the global creator audience.
“By prioritizing iOS and Android apps instead of desktop workflows, we enable creators to create compelling social media content anytime and anywhere, Indeed, by building on mobile, we’re able to prioritize ease of use and consumer-friendly features from day one.”
Alex Mashrabov, Founder, Higgsfield AI
User can pick out a video from the content library, select their selfie and Diffuse will generate a personalized character in that video’s style. They can also use Prompt Builder to create a video from scratch using text, images or video. Diffuse offers a deep level of personalization, creative control and fine-tuning.
Conclusion
While users have noted that the level of realism is lower than what SORA would offer, the mobile-first approach is a huge opportunity for creators and normal users to incorporate hyper-realistic video content on their platforms.
