GPT 6 Sol medium leads on quality, cost and agent time: 17.4 points higher, 2.3× cheaper and 1.5× faster.
Stet score
GPT 6 Sol mediumRank 7 of 14
57.695% range 49.5–65.717.4 points higher
Opus 4.8Rank 11 of 14
40.295% range 30.3–50.3
GPT 6 Sol medium solved 2 tasks that Opus 4.8 didn’t, and Opus 4.8 solved 1 that GPT 6 Sol medium didn’t. Both solved 12; neither solved 9.
stet-commit-f8ae1177Passed testsstet-commit-342d0ac0Equivalent to the merged fixstet-commit-60a0e6c2Passed testsA 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.
| Task | GPT 6 Sol medium | Opus 4.8 | Solved by |
|---|---|---|---|
stet-3169aefa | Passed tests | Passed tests | Both |
stet-89dfbc27 | Passed tests | Passed tests | Both |
stet-b28f4fe7 | Passed tests | Passed tests | Both |
stet-232b50ba | Passed tests | Passed tests | Both |
stet-5276aa3f | Passed tests | Passed tests | Both |
stet-commit-bbbbae09 | Passed tests | Passed tests | Both |
stet-15439c21 | Passed tests | Passed tests | Both |
stet-66762b4c | Passed tests | Passed tests | Both |
stet-commit-f0caded9 | Equivalent to the merged fix | Equivalent to the merged fix | Both |
stet-commit-5d766dde | Equivalent to the merged fix | Equivalent to the merged fix | Both |
stet-43493c52 | Passed tests | Passed tests | Both |
stet-6f84e978 | Passed tests | Passed tests | Both |
stet-commit-f8ae1177 | Passed tests | Failed | GPT 6 Sol medium only |
stet-2450ca2d | Failed | Failed | Neither |
stet-commit-342d0ac0 | Equivalent to the merged fix | Failed | GPT 6 Sol medium only |
stet-commit-60a0e6c2 | Failed | Passed tests | Opus 4.8 only |
stet-commit-5e075641 | Failed | Failed | Neither |
stet-commit-2826765e | Failed | Failed | Neither |
stet-commit-a8716b88 | Failed | Failed | Neither |
stet-commit-30506f8e | Failed | Failed | Neither |
stet-commit-87814583 | Failed | Failed | Neither |
stet-commit-d2ab0c25 | Failed | Failed | Neither |
stet-commit-e41c2e2d | Failed | Failed | Neither |
stet-commit-e928166f | Failed | Failed | Neither |
| Metric | GPT 6 Sol medium | Opus 4.8 |
|---|---|---|
| Rank | 7 of 14 | 11 of 14 |
| Stet score | 57.695% range 49.5–65.7 | 40.295% range 30.3–50.3 |
| Solved | 14/2411 passed, 3 judged equivalent | 13/2411 passed, 2 judged equivalent |
| Passed review | 8/24 | 8/231 not recorded |
| Matched merged change | 10/24 | 11/24 |
| Cost / task | $3.01 | $7.05 |
| How cost was recorded | estimated from tokens | reported by its harness |
| Agent time / task | 15.4 min | 23.3 min |
| Ran in | Daytona sandbox | local Docker |
The leaderboard’s readings on groups of tasks. Each group holds fewer tasks than the overall standings, so treat these as leads to test on your repo.
11 tasks
11 tasks whose merged change was 2–149 lines in 1–2 files. The 7 small fixes below are among them.
Opus 4.8 matched the merged change more often: 9 of 11 to GPT 6 Sol medium’s 6.
| GPT 6 Sol medium | Opus 4.8 | |
|---|---|---|
| Matched merged change | 6/11 | 9/11 |
| Passed review | 6/11 | 8/11 |
| Cost / task | $0.75 | $1.41 |
| Agent time / task | 6.6 min | 12.8 min |
7 tasks
7 tasks whose instructions start with “Fix”. Their merged changes were 4–29 lines in 1–2 files.
Opus 4.8 matched the merged change more often: 7 of 7 to GPT 6 Sol medium’s 5.
| GPT 6 Sol medium | Opus 4.8 | |
|---|---|---|
| Matched merged change | 5/7 | 7/7 |
| Passed review | 5/7 | 7/7 |
| Cost / task | $0.60 | $1.17 |
| Agent time / task | 5.4 min | 9.7 min |
13 tasks
13 tasks whose merged change ran over 150 lines or more than 3 files: 113–1,420 lines in 2–19 files.
GPT 6 Sol medium matched the merged change more often: 4 of 13 to Opus 4.8’s 2.
| GPT 6 Sol medium | Opus 4.8 | |
|---|---|---|
| Matched merged change | 4/13 | 2/13 |
| Passed review | 2/13 | 0/121 not recorded |
| Cost / task | $4.92 | $11.83 |
| Agent time / task | 22.9 min | 32.2 min |
Stet replays your merged PRs, scores quality above pass/fail, and delivers comparison reports through your agent.
Run it on your repocompare-n24-r3b.