Flim provides specialized generative models tailored for different creative requirements. You can select a specific model based on whether you need high-fidelity images, cinematic motion, or advanced video control.
Nano Banana 2, 2 Lite, and Pro
These models are built specifically for static image generation.
Nano Banana 2 Lite: Best for quick visual concept testing and rapid iteration during early brainstorming stages. Included with unlimited generations in the Pro Yearly Plan.
Nano Banana 2: Provides enhanced detail and improved text-prompt accuracy for refining your visual ideas.
Nano Banana Pro: The premium image generation model on Flim. Use this model for high-fidelity treatments, complex compositions, and presentation-ready concept art up to 4K resolution.
Veo 3.1
This model is designed for cinematic text-to-video generation.
When to use: Select Veo 3.1 when you need to generate high-quality video clips directly from descriptive text prompts.
Capabilities: It delivers strong aesthetic coherence, cinematic lighting, and realistic movement, making it ideal for creating high-resolution video references between 4 and 15 seconds long.
Seedance 2.0
This model excels at multi-reference image-to-video generation, allowing you to combine multiple visual inputs into a single, coherent video.
When to use: Choose Seedance 2.0 when you want to animate one or several reference images—whether they're cinematic stills from the Flim database, AI-generated images, or your own uploaded files.
Capabilities: Upload multiple visual references to guide composition, characters, style, lighting, and motion while maintaining strong visual consistency throughout the generated sequence. Seedance also supports audio references (voice-over, music, rhythm, etc.) to further influence the final result.