Meet Minimax in GPTunneL!
Through Creative.Lab, GPTunneL now offers its first neural network for creating video – Minimax. Users can now not only generate images but also produce short video clips from text descriptions. This new tool makes it fast and simple to create unique video content.
What can Minimax do?
Minimax generates videos based on your description.
The tool stands out for:
- Realistic detail in spaces and objects;
- A maximum clip length of up to 6 seconds (as of December 2024);
- Strong handling of lighting, camera angles, and object positioning;
- The ability to create impressive videos when prompts are precisely worded;
- A wide range of supported styles: from realism to CGI and anime;
- Video quality of 1280 x 720 pixels (720p) at 25 frames per second.
How to use Minimax?
Generating a video is as simple as it gets:
- Go to the Video section in Creative.Lab on GPTunneL.
- Select Minimax among the available tools. On our platform it's listed as “Minimax”.
- Enter a text description to serve as the basis for the clip.
- Wait for processing to finish.
- Watch the finished video or download it to your device.
A video is generated in roughly one to two minutes, depending on the complexity of the scene.
Sample generation
Prompt: Create a video ad for Coffee Novi that highlights the cozy and inviting atmosphere. Show glimpses of people enjoying their coffee and chatting with friends. Focus on warm lighting, comfortable seating, and the gentle sounds of a coffee shop. End with a shot of the logo and a call to action like 'Visit Us Today' or 'Your Perfect Coffee Awaits’
The model successfully carried out the task described in the prompt. It depicted customers enjoying coffee and chatting with each other in detail. Minimax also struck a good balance between light and shadow and handled the positioning of people and objects well. At the end of the video, as requested, the model displayed the Coffee Novi logo with a call to action, but did so creatively, working it into a pattern on a coffee cup.
That said, the model isn't without flaws. Some "drifting" of objects and characters is noticeable in the background. Faces and objects occasionally look distorted, and cups appear where they shouldn't. Defects are also visible in the "Your Perfect Coffee Awaits" text under the logo.
Overall, the neural network already produces impressive short clips, but developers still have work to do to refine it.
Bring images to life with Minimax!
We've added the ability to animate images with Minimax!
- Upload a photo
- Enter the prompt you want
- Watch a still image turn into motion.
Sample generation
Animation prompt: An old woman holding a sign suddenly blinks, raises one eyebrow, and gives a playful smile to the camera.
We made a picture move! Just one prompt, and a static photo came to life: a blink, a raised eyebrow, a playful smile — and there's already a connection with the viewer. No unnecessary motion, no overload — just emotion, focus on the face, and an unexpected sense of presence. This video speaks for itself: realism is preserved, attention is captured, the viewer is engaged.
Once generation is complete, you can:
- Download it to your device.
- Copy the link.
- Delete it if you no longer need it.
- View it in your personal gallery on the platform.
Try Minimax today!
Check out the capabilities of the first neural network for video creation and photo animation in GPTunneL. Head to Minimax and start generating unique content right now!