aivs.tv

Same model over timeSingle-run anecdoteAI Coding

Stickman lightsaber fight rerun, Oct 1 vs Oct 20 — Claude Opus 5.5 (Oct 1) vs Claude Opus 5.5 (Oct 20)

0:00 / 0:15
Posted byDaniel Zambrini@DanielZambrini · on X
View post on X

58 likes14,574 views

Scoreboard2

  1. Claude Opus 5.5Anthropic
    • Effort Max
    • Written as “Opus 5.5”
    • Labelled “01 Oct 2026 · Claude Opus 5.5 · Max”, US$12, 1h16m
  2. Claude Opus 5.5Anthropic
    • Effort Max
    • Written as “Opus 5.5”
    • Labelled “20 Oct 2026 · Claude Opus 5.5 · Max”, US$18.88, 1h16m

Prompt

Stickman lightsaber fight rerun, Oct 1 vs Oct 20The post doesn’t quote the prompt wording.

Author’s verdict
They say the Oct 20 rerun is as good as launch, maybe better, and the only big difference is that it used more tokens.

@DanielZambrini’s view (paraphrased)

Claude Opus 5.5 at Max reruns the author’s stickman lightsaber fight test on a lava bridge on Oct 1 and again on Oct 20, cut back and forth. The author asks whether Opus 5.5 was really nerfed.

Same model, same prompt, run on different dates: one sample each, not a controlled test. Model outputs vary from run to run, so the poster’s claim is not verified here. All same-model-over-time posts.

The post, itemised

Kind
Side-by-side
Task
Stickman lightsaber fight rerun, Oct 1 vs Oct 20
Mode
Not stated
Posted
Video length
15 s
Likes
58
Views
14,574
Bookmarks
34
Sponsored
No

Cost & time, as stated in the postFrom the video labels: Oct 1 run US$12, 1h16m; Oct 20 run US$18.88, 1h16m. The author says the rerun used more tokens.

Related

Claude Opus 5.5

Posted by @DanielZambrini on X. aivs.tv claims no ownership: it hosts a copy for fast playback and links to the post on X. Request removal

More with these models