Skip to the comparison
STET
Run it on your repo

Grok 4.7 high vs Opus 4.8 medium

Grok 4.7 high leads on quality, speed and cost.

Quality
+21.5 pts
Grok 4.7 high
14–4 head to head, 6 tied
Not yet conclusive across all 91 comparisons on the board.
Speed
1.2× faster
Grok 4.7 high
18.9 min vs 23.3 min per task
Cost
1.2× cheaper
Grok 4.7 high
$5.76 vs $7.05 per task

24 tasksData from

Stet score

Grok 4.7 highcursor · high reasoningRank 4 of 14

60.795% range 50.1–70.221.5 points higher

Opus 4.8 mediumclaude-code · medium reasoningRank 10 of 14

39.295% range 29.0–49.7

Their 95% score ranges don’t overlap.

Quality

Head-to-head matchups

Grok 4.7 high won14

Tied6

Opus 4.8 medium won4

Grok 4.7 high’s win share 69%, 95% range 52–84%Ties are within judge noise.

Task by task

Grok 4.7 high solved 2 tasks that Opus 4.8 medium didn’t, and Opus 4.8 medium solved 2 that Grok 4.7 high didn’t. Both solved 11; neither solved 9.

On the 11 tasks both solved, judges preferred Grok 4.7 high’s patch on 8, Opus 4.8 medium’s on 1, and called 2 even.

stet-43493c52Solved by 9 of 14 setups

  • Grok 4.7 highFailed
  • Opus 4.8 mediumPassed tests

Judges preferred Opus 4.8 medium’s patch

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
TaskGrok 4.7 highOpus 4.8 mediumSolved byBetter patch
stet-3169aefaPassed testsPassed testsBothGrok 4.7 high
stet-89dfbc27Passed testsPassed testsBothGrok 4.7 high
stet-b28f4fe7Passed testsPassed testsBothGrok 4.7 high
stet-232b50baPassed testsPassed testsBothGrok 4.7 high
stet-5276aa3fPassed testsPassed testsBothEven
stet-commit-bbbbae09Passed testsPassed testsBothGrok 4.7 high
stet-15439c21Passed testsPassed testsBothGrok 4.7 high
stet-66762b4cPassed testsPassed testsBothGrok 4.7 high
stet-commit-f0caded9Equivalent to the merged fixEquivalent to the merged fixBothEven
stet-commit-5d766ddeEquivalent to the merged fixEquivalent to the merged fixBothGrok 4.7 high
stet-43493c52FailedPassed testsOpus 4.8 medium onlyOpus 4.8 medium
stet-6f84e978Passed testsPassed testsBothOpus 4.8 medium
stet-commit-f8ae1177Passed testsFailedGrok 4.7 high onlyGrok 4.7 high
stet-2450ca2dPassed testsFailedGrok 4.7 high onlyGrok 4.7 high
stet-commit-342d0ac0FailedFailedNeitherEven
stet-commit-60a0e6c2FailedPassed testsOpus 4.8 medium onlyOpus 4.8 medium
stet-commit-5e075641FailedFailedNeitherGrok 4.7 high
stet-commit-2826765eFailedFailedNeitherEven
stet-commit-a8716b88FailedFailedNeitherEven
stet-commit-30506f8eFailedFailedNeitherGrok 4.7 high
stet-commit-87814583FailedFailedNeitherGrok 4.7 high
stet-commit-d2ab0c25FailedFailedNeitherGrok 4.7 high
stet-commit-e41c2e2dFailedFailedNeitherOpus 4.8 medium
stet-commit-e928166fFailedFailedNeitherEven

Side by side

Side by side
MetricGrok 4.7 highOpus 4.8 medium
Harnesscursorclaude-code
Reasoning efforthighmedium
Rank4 of 1410 of 14
Stet score60.795% range 50.1–70.239.295% range 29.0–49.7
Solved13/2411 passed, 2 judged equivalent13/2411 passed, 2 judged equivalent
Cost / task$5.76$7.05
Agent time / task18.9 min23.3 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 14 setups: bmk_d4ba310fdf50db4ccb2092e3.

See all 14 setups on the leaderboardHow Stet measures