# Compare four AI video editors with the same footage

Use a shared brief, a limited revision budget and separate checks for the export and editable project.

[HTML](https://lukejbyrne.com/library/guides/q_dky8brzhy/)
Content language: en
Author: Luke Byrne
Video published: 2026-10-06

[Watch the video](https://www.youtube.com/watch?v=q_DKy8BrZhY)
Video language: English

- [Download the free AI Video Editor skill ZIP](https://lukejbyrne.com/library/downloads/ai-video-editor.zip)

## Define the comparison before running it

Luke supplies the same raw footage and saved editing Skill to Opus 5.5, Fable 5.1, Astra and Sol. Each gets two follow-up interactions.

1. Keep identical source footage and the same initial instructions.
2. Define how you will judge pauses, bad takes, explanation and graphics.
3. Record the follow-ups and delivery format for each run.

## Inspect the edits as complete videos

Luke prefers Opus overall, while Fable produces some strong individual graphics. The GPT edits cut the footage but their visuals often feel like animated slides to him.

1. Watch each opening, explanation and ending in sequence.
2. Check whether cuts preserve the meaning and whether a graphic actually explains the narration.
3. Use each follow-up for a specific defect, then judge the revised export against the same criteria.

## Check the project and interpret the timing

Some outputs are MP4s, others include editable Final Cut projects. Elapsed time includes waiting for Luke and shared-computer work.

1. Open the editable project when supplied and check source links and timeline structure.
2. Keep editability separate from visual quality in the comparison.
3. Record active work, waiting and rendering separately where possible.
4. Treat Luke’s preference as the outcome of this workflow test, not a universal ranking of models.

## The takeaway

A fair comparison needs the same task and visible revision history. Different deliverables and shared-computer timing limit the conclusions.

## Workflow resources


## Chapters

- [0:00 Same footage, four AI editors](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=0s)
- [0:31 The raw footage and test setup](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=31s)
- [1:17 My video editing skill and judging criteria](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=77s)
- [2:14 GPT-6 Astra results](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=134s)
- [3:04 GPT-6.1 Sol results](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=184s)
- [3:47 Claude Opus 5.5 results](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=227s)
- [5:39 Claude Fable 5.1 results](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=339s)
- [7:46 Follow-up prompts and rendering](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=466s)
- [9:05 Time taken and editable projects](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=545s)
- [9:58 My verdict and editing costs](https://www.youtube.com/watch?v=q_DKy8BrZhY&t=598s)

## Sources and limitations

Written follow-along guide prepared 31 October 2026 from this exact published video’s description and chapter sequence. The checklists are editorial adaptations, not verbatim transcript quotations. Demonstrated results are attributed to the video; the full final video was not independently transcribed or watched for this article, and the product was not retested. For installation commands, later interface changes and current offers, use the original resources linked here.

[Original video](https://www.youtube.com/watch?v=q_DKy8BrZhY)