Skip to the comparison
STET
Join the waitlist

GLM 5.3 Flash high vs Haiku 5.5 max

No clear quality winner: Haiku 5.5 max won 15, GLM 5.3 Flash high won 5, 3 tied, 1 undecided; GLM 5.3 Flash high leads on cost and agent time: 9.4× cheaper and 2.2× faster.

24 tasksData from

Stet score

GLM 5.3 Flash highcodex · high reasoningRank 15 of 16

29.095% range 19.9–39.3

Haiku 5.5 maxclaude-code · max reasoningRank 8 of 16

55.595% range 44.6–65.726.5 points higher

Their 95% score ranges don’t overlap.

Cost / task

GLM 5.3 Flash high$0.419.4× cheaper
Haiku 5.5 max$3.85

Agent time / task

GLM 5.3 Flash high26.7 min2.2× faster
Haiku 5.5 max58.4 min

Quality

Head-to-head matchups

GLM 5.3 Flash high won5

Tied31 undecided

Haiku 5.5 max won15

Haiku 5.5 max’s win share 71%, 95% range 53–87%Ties are within judge noise.

Task by task

GLM 5.3 Flash high solved 2 tasks that Haiku 5.5 max didn’t, and Haiku 5.5 max solved 5 that GLM 5.3 Flash high didn’t. Both solved 7; neither solved 9.

GLM 5.3 Flash high only 2

  • stet-232b50baPassed tests
  • stet-66762b4cPassed tests

Haiku 5.5 max only 5

  • stet-commit-5d766ddeEquivalent to the merged fix
  • stet-6f84e978Equivalent to the merged fix
  • stet-2450ca2dPassed tests
  • stet-commit-342d0ac0Equivalent to the merged fix
  • stet-commit-f8ae1177Passed tests

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
TaskGLM 5.3 Flash highHaiku 5.5 maxSolved by
stet-3169aefaPassed testsPassed testsBoth
stet-89dfbc27Passed testsPassed testsBoth
stet-b28f4fe7Passed testsPassed testsBoth
stet-232b50baPassed testsFailedGLM 5.3 Flash high only
stet-5276aa3fPassed testsPassed testsBoth
stet-commit-bbbbae09Passed testsPassed testsBoth
stet-66762b4cPassed testsFailedGLM 5.3 Flash high only
stet-15439c21Passed testsPassed testsBoth
stet-commit-5d766ddeFailedEquivalent to the merged fixHaiku 5.5 max only
stet-43493c52FailedFailedNeither
stet-commit-f0caded9Passed testsPassed testsBoth
stet-6f84e978FailedEquivalent to the merged fixHaiku 5.5 max only
stet-2450ca2dFailedPassed testsHaiku 5.5 max only
stet-commit-342d0ac0FailedEquivalent to the merged fixHaiku 5.5 max only
stet-commit-f8ae1177FailedPassed testsHaiku 5.5 max only
stet-commit-5e075641FailedFailedNeither
stet-commit-2826765eFailedFailedNeither
stet-commit-a8716b88FailedFailedNeither
stet-commit-e928166fFailedFailedNeither
stet-commit-30506f8eFailedFailedNeither
stet-commit-87814583FailedFailedNeither
stet-commit-d2ab0c25FailedFailedNeither
stet-commit-e41c2e2dFailedFailedNeither

Side by side

Side by side
MetricGLM 5.3 Flash highHaiku 5.5 max
Harnesscodexclaude-code
Reasoning efforthighmax
Rank15 of 168 of 16
Stet score29.095% range 19.9–39.355.595% range 44.6–65.7
Solved9/249 passed, 0 judged equivalent12/249 passed, 3 judged equivalent
Cost / task$0.41$3.85
Agent time / task26.7 min58.4 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