Runway's Gen-2 best mirrors what you'd get using Open AI's Sora, using a multimodal AI system to generate video clips using text prompts.

Runway's Gen-2 capabilities include the ability to upload images or videos for use as a reference for the video clip you'd like to generate. Whether Open AI's Sora will also support creating video clips from reference images or clips remains to be seen.

Judging from the quality of Sora-generated clips shared by OpenAI, Sora bests Runway Gen-2 as an AI text-to-video generator. However, given the speed of development in the AI space (and the fact that Runway launched Gen-2 about a year before the first preview of Sora was released), it's clear OpenAI's Sora and Runway Gen-2 (and its future versions) will battle for the best text-to-video AI generator title.

2 Pika

Free: 250 initial credits (free 30 credits daily after exhausting initial credits) Standard: $10/month ($8 when paid annually) for 700 monthly credits

Pika is another AI-powered video generator that can create videos and 3D animations from text prompts and images. Pika is available on the web app and Discord. However, the platform you use determines the output quality and additional features you can access.

The web app allows you to modify specific regions in your generated clip, expand your video canvas, and add lip sync to your generated videos. These features aren't available on the Discord server option.

That said, I recommend trying out the web and Discord options to see which gives you better results. The clip below was generated on Pika's web version using the same prompt as the viral "Lady Walking in Tokyo" video by OpenAI Sora:

A stylish woman walks down a Tokyo street filled with warm glowing neon and animated city signage. She wears a black leather jacket, a long red dress, and black boots, and carries a black purse. She wears sunglasses and red lipstick. She walks confidently and casually. The street is damp and reflective, creating a mirror effect of the colorful lights. Many pedestrians walk about.

Using the same prompt (/create prompt) on Pika's Discord server gave the result below:

We'll let you judge which is better, but it's clear Pika has some catching up to do compared to the quality of Sora-generated clips online. However, its other features, like lip-syncing and image animation, give it an edge over Sora—at least for now.

3 Pixverse

Free

Pixverse is another alternative to Open AI's Sora that lets you create realistic videos with text prompts. Pixverse also offers two platforms for video creation: the web platform and the Discord server.

Pixverse's web platform provides a more comprehensive video creation experience where you can create, view, filter, and edit all the videos you generate.

The video above was generated on Pixverse's web version. While you can always regenerate to get better results (it's free!), the Discord server option has the advantage of generating four clips at a go. This gives you the option of picking which is best without regenerating multiple times. Below is a sample generated on its Discord server:

You can join Pixverse's Discord server and generate your clips using the /create command. You can also select the aspect ratio and negative prompt (if needed) for your videos.

Quality-wise, Pixverse is in the same class as Pika—below Sora.

4 Kaiber

Free: 100 initial credits Explorer: $5/month for 300 credits.

Kaiber is an artist-focused AI video generation tool that allows you to create videos from images or text descriptions.

Kaiber also supports audio reactivity, which means you can upload a song and let the AI generate a video that matches the rhythm and mood of the music. You can also customize your video's length, dimensions, camera movements, and starting frame. You can use Kaiber on the web or through its mobile apps.

The biggest allure of Kaiber is its ability to generate clips that match the rhythm of uploaded sounds. Its artist-centric features also help prop it up against Sora. However, in terms of generated clip realism, Sora still stands clear.

5 Synthesia

Free: Test video Starter: $29/month ($22 when billed annually) for 10 minutes of video.

Synthesia is an AI text-to-video generator that allows you to create realistic talking videos from text scripts. You can choose from various avatars, backgrounds, and languages to customize your video.

Synthesia differs from Sora in that it does not generate the visuals from scratch but uses existing footage and modifies it to match the text. Synthesia is also limited to talking videos, while Sora can generate any kind of video from text.

Synthesia is a good alternative to Sora for creating engaging and personalized videos for education, marketing, or entertainment purposes.

6 Vidnoz

Free: 1 minute/day Starter: $19.99 ($14.99 when billed annually) for 10 minutes of watermark-free videos

Vidnoz is another AI video generator that creates talking videos from text scripts. Vidnoz AI uses natural language processing (NLP) and computer vision to generate realistic lip-syncing and facial expressions for the avatars. You can also customize the avatars' appearance, clothing, and accessories.

Vidnoz AI is similar to Synthesia in terms of functionality, but it allows for more customization when creating your free test video. You can choose between avatars and voices, something Synthesia doesn't support.

If what followed OpenAI's launch of ChatGPT is anything to go by, then you can expect more AI text-to-video platforms to be introduced. You can also expect Google's Lumiere and Meta's Make-A-Video to become accessible to the public.