Skip to the comparison
STET
Join the waitlist

GPT 6 Luna low vs SWE-2 Max high

SWE-2 Max high leads on quality: won 17, tied 4, lost 2, 1 undecided head to head; GPT 6 Luna low leads on agent time: 9.9× faster; SWE-2 Max high’s cost wasn’t recorded.

24 tasksData from

Stet score

GPT 6 Luna lowcodex · low reasoningRank 16 of 16

19.995% range 11.6–29.5

SWE-2 Max highdevin · high reasoningRank 3 of 16

64.395% range 54.3–73.944.4 points higher

Their 95% score ranges don’t overlap.

Cost / task

GPT 6 Luna low$0.05
SWE-2 Max highnot recorded

Agent time / task

GPT 6 Luna low6.9 min9.9× faster
SWE-2 Max high68.3 min

Quality

Head-to-head matchups

GPT 6 Luna low won2

Tied41 undecided

SWE-2 Max high won17

SWE-2 Max high’s win share 83%, 95% range 69–95%Ties are within judge noise.

Task by task

GPT 6 Luna low solved 0 tasks that SWE-2 Max high didn’t, and SWE-2 Max high solved 10 that GPT 6 Luna low didn’t. Both solved 5; neither solved 8.

GPT 6 Luna low only 0

SWE-2 Max high only 10

  • stet-5276aa3fPassed tests
  • stet-commit-bbbbae09Passed tests
  • stet-66762b4cPassed tests
  • stet-commit-5d766ddePassed tests
  • stet-commit-f0caded9Passed tests
  • stet-6f84e978Passed tests
  • stet-2450ca2dPassed tests
  • stet-commit-342d0ac0Equivalent to the merged fix
  • stet-commit-5e075641Passed tests
  • stet-commit-a8716b88Equivalent 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 lowSWE-2 Max highSolved by
stet-3169aefaPassed testsPassed testsBoth
stet-89dfbc27Passed testsPassed testsBoth
stet-b28f4fe7Passed testsPassed testsBoth
stet-232b50baPassed testsPassed testsBoth
stet-5276aa3fFailedPassed testsSWE-2 Max high only
stet-commit-bbbbae09No change madePassed testsSWE-2 Max high only
stet-66762b4cFailedPassed testsSWE-2 Max high only
stet-15439c21Passed testsEquivalent to the merged fixBoth
stet-commit-5d766ddeNo change madePassed testsSWE-2 Max high only
stet-43493c52FailedFailedNeither
stet-commit-f0caded9FailedPassed testsSWE-2 Max high only
stet-6f84e978FailedPassed testsSWE-2 Max high only
stet-2450ca2dFailedPassed testsSWE-2 Max high only
stet-commit-342d0ac0FailedEquivalent to the merged fixSWE-2 Max high only
stet-commit-f8ae1177No change madeFailedNeither
stet-commit-5e075641FailedPassed testsSWE-2 Max high only
stet-commit-2826765eFailedFailedNeither
stet-commit-a8716b88FailedEquivalent to the merged fixSWE-2 Max high only
stet-commit-e928166fFailedFailedNeither
stet-commit-30506f8eFailedFailedNeither
stet-commit-87814583FailedFailedNeither
stet-commit-d2ab0c25FailedFailedNeither
stet-commit-e41c2e2dFailedFailedNeither

Side by side

Side by side
MetricGPT 6 Luna lowSWE-2 Max high
Harnesscodexdevin
Reasoning effortlowhigh
Rank16 of 163 of 16
Stet score19.995% range 11.6–29.564.395% range 54.3–73.9
Solved5/245 passed, 0 judged equivalent16/2413 passed, 3 judged equivalent
Cost / task$0.05not recorded
Agent time / task6.9 min68.3 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