Verified specification snapshot
| Field | Published information |
|---|---|
| Provider | Luma |
| Single-pass generation | Not confirmed in the cited source |
| Resolution | 1080p; HDR and EXR workflow documented |
| Audio | Native synchronized audio not verified |
| Direction and references | Up to 16 keyframes; Modify Video V2 |
| Access | Luma App and API |
| Billing | Credits vary by model and output settings |
Important limit: The 20-second figure refers to footage modification. We do not present it as a verified text-to-video single-pass generation limit.
Choose a footage-editing task with a clear invariant
For an existing shot, decide what must remain fixed before you change its appearance. That may be the actor’s timing, camera path or position of a product. Write the invariant down and keep an untouched copy of the original footage for a side-by-side review.
Try a bounded transformation first: change the setting while preserving the subject’s motion, or change one surface while keeping the rest of the frame. If the original contains a complex occlusion, inspect that moment carefully in the transformed clip. A visually attractive still frame does not establish a successful video edit.
Make HDR and EXR a delivery decision
Luma’s official information describes HDR generation and EXR export for grading and compositing workflows. Those features are useful only if the receiving pipeline can interpret them correctly. Agree on the working color space and handoff format with the person doing the grade before rendering a whole batch.
Test one clip through the actual editor or compositor, then export a review version that your reviewer can see correctly. Compare highlights, shadow detail and skin tones across the round trip. Do not infer a complete color-managed pipeline from a label in a model selector.
Why the duration column is deliberately unconfirmed
The source explicitly gives a maximum input length for Modify Video V2. It does not, in the information used here, establish a matching text-to-video single-pass limit. We leave that comparison field unconfirmed instead of substituting the editing figure or carrying a number over from another version.
For your own project, record whether each duration means generated output, source footage, an extension or a sequence assembled from several shots. This simple distinction prevents misleading comparisons and helps you estimate real production work. For a long scene, budget for both generation and the editing needed to connect its parts.
Decide using your own accepted shots
Our workflow recommendation is an editorial inference from the documented controls. We have not run a controlled benchmark or purchased a test batch for this guide. Compare the same required scene across the tools you can actually access, then record acceptance rate, edit time and total cost.
Keep a usable shot only when it meets the brief at delivery size. A model with a smaller advertised limit can still be a practical choice for a short project. For a fair test, keep the required subject, action and framing consistent while adapting syntax to each tool’s documented controls.
Compare the task, then the alternative
Runway’s Gen-4.5 help page documents short text- and image-led generation at 720p base output. Source ↗ Google’s Omni guide describes conversational changes and explicitly labels higher-resolution exports as upscaled. Source ↗
These are different production capabilities, not evidence of a quality winner. Add an alternative to your shortlist only when it supports an important requirement of your shot. Use the directory to examine the details before starting a paid pilot.
Sources and verification
Reviewed October 25, 2026. Provider documentation supports specifications; it is not evidence of an independent performance test.