Claude Fable 5.2 vs Opus 5.5 for video editing: either will do. We had both cut the same talk, and the two shorts are nearly the same video. On Fable 5.2, leave the effort at Medium, the default. If you cut several videos a week on a subscription, Opus 5.5 goes further.
The talk was an 11-and-a-half-minute startup talk. Fable 5.2 got one sentence, "cut this into a motion-graphics TikTok," and twelve minutes later we had a finished short. Opus 5.5 cut the same talk in the same Motion graphics style, and so did Fable 5.2 at High effort. Here are all three, in this order: Opus 5.5, Fable 5.2 at Medium (the Claude Code default), Fable 5.2 at High.
They are nearly the same video
We expected a model that costs twice as much, or a higher effort setting, to make a visibly different video. Then we put the three side by side and had to check which was which.
Each was cut on its own, with nobody stepping in. They still made the same calls:
- None opens where the talk opens. All three pulled up the same line from near the end, "The worst failure mode isn't being wrong. It's not making a decision," and set it under the same title.
- The speaker says an early founder is in fog and can see 10 feet ahead. All three drew fog with a 10-foot circle in it.
- He says walking fast in one direction gets you far more information. All three drew bars labeled "Information" and filled them up.
- All three end on a card that reads "Pick one. Go deep."
Where they differ is layout and color. Opus 5.5 put the diagram on top and the speaker below, on cream cards. Fable 5.2 at Medium tucked the speaker into a window at the top and went dark with orange and yellow. At High it let the map fill everything below the speaker and set the captions on a plate.
Each also kept a line the others dropped. Medium kept "burn the other boats." High went back to early in the talk for "you can only figure out what you should be working on by making contact with reality and getting feedback from customers," drew a shaft sunk all the way down to "reality" for it, and used that shaft again on the closing card.
Time and cost
| Opus 5.5 | Fable 5.2, Medium | Fable 5.2, High | |
|---|---|---|---|
| From request to finished short | 30 min 30 s | 12 min 11 s | 34 min 33 s |
| Trips to the model | 79 | 22 | 65 |
| Output tokens | About 113,000 | About 51,000 | About 157,000 |
| Cost at API list prices | About $6.00 | About $4.60 | About $17 |
| Length of the short | 0:45 | 0:53 | 0:51 |
Start with the two Fable 5.2 columns. Same model, same sentence, same file. High took nearly three times as long and cost more than three times as much, for a short that thought one step further and still looks like the same video at a glance.
Then Opus 5.5. Per token, Fable 5.2 costs twice as much, but at Medium it made far fewer trips, so this one video came out cheaper. On a Claude subscription you pay none of these figures, and the edit counts against your usage.
We cut the talk once with each, and the three were not set up exactly alike, so take these numbers as rough. What we can say is that Fable 5.2, at its default effort, took one sentence and came back in twelve minutes, with nobody stepping in, holding a short we would post.
Fable 5.2 or Opus 5.5: which to use
The videos are nearly the same, so choose by time and usage.
Leave the effort at Medium, the default. The High cut did think one step further, but the extra time and usage did not buy a clearly better video.
On a subscription, mind the limit. On Pro plans and standard Team seats, Fable can use only half of your weekly usage. If you cut several videos a week, Opus 5.5 goes further. Anthropic's own advice is to start with Opus 5.5 as well.
On the API, work out the cost per video. Per token Fable 5.2 is double. For this video, Medium came out cheaper than Opus 5.5 and High far dearer.
Whichever you choose, type /model in Claude Code before you start, pick one, and let it finish the edit.
A longer recording, a take full of restarts, or a style that generates a lot of images may pull the models and effort levels apart. We haven't tested that yet, and we will add it here when we do.
What is new in Claude Fable 5.2
If you skipped the announcement, these are Anthropic's numbers.
| Fable 5.2 | For comparison | |
|---|---|---|
| Released | November 9, 2026 | Open to all users |
| API price per million tokens | $8 in, $40 out | Fable 5.1 was $10 and $50. Opus 5.5 is $4 and $20 |
| Speed | Output up to 30% faster than Fable 5.1 | Fast mode is up to 2.5 times faster, at $16 and $80 |
| Reading charts and figures (Chartography) | 73.4% | Fable 5.1 scored 44.8%, Opus 5.5 64.4% |
| In Claude Code | Medium effort by default | |
| Pro plans and standard Team seats | Included for up to 50% of weekly usage limits |
The chart-reading row has something to do with video. Claude looks at its own exported frames to check the result, and that depends on how well it reads a picture. The test covers charts and figures, though, and says nothing about video frames, so treat it as a hint.
Try it
- Open a style on Claude video editing skills, copy its install line and paste it into Claude Code.
- Type
/modeland choose Fable 5.2 or Opus 5.5. Leave the effort alone, or change it with/effort. - Ask for the video, for example: "Cut ~/Movies/explainer.mov into a motion-graphics TikTok."
The skills are free, and so are the music, sound effects and stock footage. The transcript uses Pireel credits by the length of your video, and a new account gets 100 credits. Claude Code also needs ffmpeg and Google Chrome, and Claude sets them up with you the first time.
If you would rather edit by deleting words from a transcript yourself, see text-based editing vs AI agent editing.
Put this into practice
Frequently asked questions
Can Claude Fable 5.2 edit videos?
Yes, from Claude Code. Claude works on the video files on your computer with tools such as ffmpeg, and a skill tells it how to pick the passage, cut it, design the graphics and check the export.
Fable 5.2 or Opus 5.5 for video editing?
Either will do. We had both models cut the same talk and the two shorts were nearly the same video. Fable 5.2 finished in 12 minutes against 30 for Opus 5.5. Each model cut it once, so treat that gap as a hint.
Should I run Fable 5.2 at Medium or High effort for video editing?
Medium, the default, is enough. We cut the same talk at both: Medium took 12 minutes and High about 35, at more than three times the cost at API list prices, for nearly the same video.
How much does Claude Fable 5.2 cost?
On the API, $8 per million input tokens and $40 per million output tokens; fast mode is $16 and $80. On Pro plans and standard Team seats, Fable is included for up to 50% of the weekly usage limit.
Is my video uploaded when Claude edits it?
No. With Pireel's skills, only the sound goes to Pireel to make the transcript. The picture, the edit and the finished file stay on your computer.


