Skip to the comparison
STET
Get a free trial

Haiku 5.5 low+guidance vs SWE-2 Max

SWE-2 Max leads on quality: 24.3 points higher; Haiku 5.5 low+guidance leads on agent time: 5.5× faster; SWE-2 Max’s cost wasn’t recorded.

24 tasksData from

Stet score

Haiku 5.5 low+guidanceclaude-code · low reasoningRank 10 of 15

42.495% range 33.2–51.7

SWE-2 Maxdevin · high reasoningRank 1 of 15

66.795% range 56.8–75.924.3 points higher

Their 95% score ranges don’t overlap.

Cost / task

Haiku 5.5 low+guidance$0.29
SWE-2 Maxnot recorded

Agent time / task

Haiku 5.5 low+guidance12.4 min5.5× faster
SWE-2 Max68.3 min

Quality

Head-to-head matchups

Haiku 5.5 low+guidance won4

Tied5

SWE-2 Max won15

SWE-2 Max’s win share 71%, 95% range 54–86%Ties are within judge noise.

Task by task

Haiku 5.5 low+guidance solved 1 task that SWE-2 Max didn’t, and SWE-2 Max solved 8 that Haiku 5.5 low+guidance didn’t. Both solved 8; neither solved 7.

Haiku 5.5 low+guidance only 1

  • stet-43493c52Passed tests

SWE-2 Max only 8

  • stet-15439c21Equivalent to the merged fix
  • stet-commit-5d766ddePassed tests
  • stet-6f84e978Passed tests
  • stet-2450ca2dPassed tests
  • stet-commit-342d0ac0Equivalent to the merged fix
  • stet-commit-60a0e6c2Passed tests
  • 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.

All 24 compared tasks
Task by task
TaskHaiku 5.5 low+guidanceSWE-2 MaxSolved by
stet-3169aefaPassed testsPassed testsBoth
stet-89dfbc27Passed testsPassed testsBoth
stet-b28f4fe7Passed testsPassed testsBoth
stet-232b50baPassed testsPassed testsBoth
stet-5276aa3fPassed testsPassed testsBoth
stet-commit-bbbbae09Passed testsPassed testsBoth
stet-66762b4cPassed testsPassed testsBoth
stet-15439c21FailedEquivalent to the merged fixSWE-2 Max only
stet-commit-f0caded9Equivalent to the merged fixPassed testsBoth
stet-commit-5d766ddeFailedPassed testsSWE-2 Max only
stet-43493c52Passed testsFailedHaiku 5.5 low+guidance only
stet-6f84e978FailedPassed testsSWE-2 Max only
stet-commit-f8ae1177FailedFailedNeither
stet-2450ca2dFailedPassed testsSWE-2 Max only
stet-commit-342d0ac0FailedEquivalent to the merged fixSWE-2 Max only
stet-commit-60a0e6c2FailedPassed testsSWE-2 Max only
stet-commit-5e075641FailedPassed testsSWE-2 Max only
stet-commit-2826765eFailedFailedNeither
stet-commit-a8716b88FailedEquivalent to the merged fixSWE-2 Max only
stet-commit-30506f8eFailedFailedNeither
stet-commit-87814583FailedFailedNeither
stet-commit-d2ab0c25FailedFailedNeither
stet-commit-e41c2e2dFailedFailedNeither
stet-commit-e928166fFailedFailedNeither

Side by side

Side by side
MetricHaiku 5.5 low+guidanceSWE-2 Max
Harnessclaude-codedevin
Reasoning effortlowhigh
Rank10 of 151 of 15
Stet score42.495% range 33.2–51.766.795% range 56.8–75.9
Solved9/248 passed, 1 judged equivalent16/2413 passed, 3 judged equivalent
Cost / task$0.29not recorded
Agent time / task12.4 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 15 setups: bmk_d4ba310fdf50db4ccb2092e3.

See all 15 setups on the leaderboardHow Stet measures