Generative image and video models in Descript

Descript lets you choose which AI model powers your generated images and videos. Different models are better at different things: some work faster, others create more polished results, and some handle complex instructions better. This article explains the available generative media models in plain language, so you can quickly decide, “Which model should I use?” without needing to be an AI expert.

This article covers

Where to pick a model
Image models at a glance
Video models at a glance
FAQ

Availability & usage

Model availability depends on your plan. Some premium models use more AI Credits than others. You can switch models anytime; choosing a new model only affects generations made after the switch.

Where to pick a model

Choose a model for each type of generative media in your App Settings, under the General tab, in the Advanced section.

Clicking this link will take you directly to your App Settings in Descript

If you only need to change the model for a single image or video generation and don't want to modify your overall account settings, you can select a model for the specific generation you are performing while using the associated AI tool.

Changing the model in this menu does not change your App Settings.

Image models at a glance

Model	Overview	Strengths	Trade-offs
Flux 2 Pro (defaut)	Balances speed and quality, with improved text rendering and restlying	Enhanced text rendering Improved restyling capabilities High-quality output	Slightly slower generation time than Flux Kontext
Flux Kontext [pro]	Well-balanced model that handles most image generation tasks with consistent, high-quality results	Context-aware edits Maintains visual consistency Balances speed and quality	Uses more credits than Flux [dev]
Flux [dev]	Fast, budget-friendly option that excels at creative and fantasy-style imagery	Quick generation Good for fantasy & illustration Low cost	Optimized for speed over fine details
GPT Image 1	Advanced model designed for complex editing tasks and working with multiple image inputs	Follows nuanced instructions Text + multi-image inputs (via Underlord) Strong editing capabilities	Prioritizes capability over speed and cost
Qwen	Specialized model engineered for text rendering and multilingual content	Text in images Chinese, Japanese, English Multi-line layouts	Focused on text rather than artistic imagery
Nano Banana	Versatile model that provides reliable results across a wide range of image tasks	Speed-to-quality balance Precise image editing Great at complex prompts	Other models may be better at highly specialized tasks
Nano Banana Pro	Advanced model offering enhanced performance for creative and technical image tasks	Higher fidelity and realism Improved lighting and texture rendering Handles artistic and photorealistic prompts equally well	May be slower than standard Nano Banana

Video models at a glance

Model	Overview	Strengths	Trade-offs
Pixverse v5 (default)	Well-balanced model for most video generation needs with reliable quality and performance	Smooth motion & stable camera Good lighting & atmosphere Speed–quality–cost balance	Use specialized models for advanced physics or premium cinematic quality
Hailuo 02	Physics-focused model that excels at realistic movement, character consistency, and complex actions	Exceptional physics simulation Character consistency Complex movements & acrobatics	Prioritizes realism over speed and cost efficiency
Kling 01	Next-generation cinematic model with extended video length and enhanced visual quality	Generates up to 10 seconds Latest cinematic rendering technology High-resolution output	Higher credit usage per second of video
Kling Video 2.5 Turbo Pro	Premium model engineered for professional-quality cinematic video with exceptional detail and motion	Movie-quality output Exceptional motion fluidity Strong character animation High-resolution 1080p	Longer generation times and higher credit usage
Veo 3.1	Top-tier model combining highest visual quality with integrated audio for complete video production	High-fidelity video Synchronized dialogue & sound Professional cinematic output	Designed for final production rather than rapid iteration
Veo 3.1 [fast]	Streamlined version of Veo optimized for faster turnaround while maintaining strong visual quality	Much faster than Veo Similar visual quality Cost-effective iteration	Some advanced features are simplified for speed
Wan v2.2 [turbo]	Speed-optimized model for rapid experimentation and quick concept validation	Very fast generation Efficient resource usage Good for concept testing	Optimized for speed and experimentation over final production quality
Sora 2	Advanced cinematic model that delivers exceptional realism with realistic physics, synchronized audio, and consistent multi-shot sequences	Realistic human faces and expressions Accurate physics simulation Synchronized dialogue and sound effects Multi-shot consistency	Cannot accept image inputs with human faces (can generate humans from scratch); premium model with higher credit usage

FAQ

Which model should I use?

It depends! The table above can help match your goal (speed vs polish vs complexity) to an available model. You can always choose a different model in your App settings and re-create your content to see what works best for you.

Do premium models use more AI Credits?

Often, yes. Higher-end models and longer videos usually consume more credits. If you’re exploring or drafting, try a “fast/lean” model first; switch to a premium model when you’re ready to finalize. Learn more about how to track and understand your Media Minutes and AI Credits.

If I change models, will my previous generations update?

No. Model changes only affect new generations. Existing images/videos stay as-is.

Why don’t I see a specific model?

Availability can vary by plan. If a model isn't available from your App Settings, check your plan details.

I'm seeing an error message: What does that mean?

These errors come directly from our model providers and can occur for a few different reasons. See Troubleshooting AI image and video generation errors.