Skip to the comparison
STET
Join the waitlist

GPT 6 Luna low vs Opus 5 medium

No clear quality winner: Opus 5 medium won 18, GPT 6 Luna low won 2, 3 tied, 1 undecided; GPT 6 Luna low leads on cost and agent time: 127.0× cheaper and 3.4× faster.

24 tasksData from

Stet score

GPT 6 Luna lowcodex · low reasoningRank 16 of 16

19.995% range 11.6–29.5

Opus 5 mediummedium reasoningRank 5 of 16

62.395% range 52.0–72.042.4 points higher

Their 95% score ranges don’t overlap.

Cost / task

GPT 6 Luna low$0.05127.0× cheaper
Opus 5 medium$6.78

Agent time / task

GPT 6 Luna low6.9 min3.4× faster
Opus 5 medium23.2 min

Quality

Head-to-head matchups

GPT 6 Luna low won2

Tied31 undecided

Opus 5 medium won18

Opus 5 medium’s win share 82%, 95% range 68–95%Ties are within judge noise.

Task by task

GPT 6 Luna low solved 0 tasks that Opus 5 medium didn’t, and Opus 5 medium solved 9 that GPT 6 Luna low didn’t. Both solved 5; neither solved 9.

GPT 6 Luna low only 0

Opus 5 medium only 9

  • stet-5276aa3fPassed tests
  • stet-commit-bbbbae09Passed tests
  • stet-66762b4cPassed tests
  • stet-commit-5d766ddeEquivalent to the merged fix
  • stet-43493c52Passed tests
  • stet-commit-f0caded9Passed tests
  • stet-6f84e978Passed tests
  • stet-2450ca2dPassed tests
  • stet-commit-2826765eEquivalent to the merged fix

A task counts as solved when its tests passed or a judge found the fix equivalent to the merged change, as in the leaderboard’s Solved column. Stet score also weighs code quality, so it can order the two differently. 1 of the 24 tasks is left out because one of the two has no scored result on it.

All 23 compared tasks
Task by task
TaskGPT 6 Luna lowOpus 5 mediumSolved by
stet-3169aefaPassed testsPassed testsBoth
stet-89dfbc27Passed testsPassed testsBoth
stet-b28f4fe7Passed testsPassed testsBoth
stet-232b50baPassed testsPassed testsBoth
stet-5276aa3fFailedPassed testsOpus 5 medium only
stet-commit-bbbbae09No change madePassed testsOpus 5 medium only
stet-66762b4cFailedPassed testsOpus 5 medium only
stet-15439c21Passed testsPassed testsBoth
stet-commit-5d766ddeNo change madeEquivalent to the merged fixOpus 5 medium only
stet-43493c52FailedPassed testsOpus 5 medium only
stet-commit-f0caded9FailedPassed testsOpus 5 medium only
stet-6f84e978FailedPassed testsOpus 5 medium only
stet-2450ca2dFailedPassed testsOpus 5 medium only
stet-commit-342d0ac0FailedFailedNeither
stet-commit-f8ae1177No change madeFailedNeither
stet-commit-5e075641FailedFailedNeither
stet-commit-2826765eFailedEquivalent to the merged fixOpus 5 medium only
stet-commit-a8716b88FailedFailedNeither
stet-commit-e928166fFailedFailedNeither
stet-commit-30506f8eFailedFailedNeither
stet-commit-87814583FailedFailedNeither
stet-commit-d2ab0c25FailedFailedNeither
stet-commit-e41c2e2dFailedFailedNeither

Side by side

Side by side
MetricGPT 6 Luna lowOpus 5 medium
Harnesscodex—
Reasoning effortlowmedium
Rank16 of 165 of 16
Stet score19.995% range 11.6–29.562.395% range 52.0–72.0
Solved5/245 passed, 0 judged equivalent14/2412 passed, 2 judged equivalent
Cost / task$0.05$6.78
Agent time / task6.9 min23.2 min

See which setup wins on your repo

Stet replays your merged PRs, scores quality above pass/fail, and delivers comparison reports through your agent.

Run it on your repo

About this data

  • 24 tasks, each a merged change replayed from the repo as it was just before the merge.
  • Data from .
  • Both setups come from the same leaderboard run of 16 setups: bmk_d4ba310fdf50db4ccb2092e3.

See all 16 setups on the leaderboardHow Stet measures