GPT 6 Luna max leads on quality and cost: 19.0 points higher and 1.1× cheaper; GLM 5.3 Flash leads on agent time: 2.2× faster.
Stet score
GLM 5.3 FlashRank 13 of 14
31.095% range 21.5–41.5
GPT 6 Luna maxRank 8 of 14
50.095% range 40.1–59.319.0 points higher
GLM 5.3 Flash solved 1 task that GPT 6 Luna max didn’t, and GPT 6 Luna max solved 4 that GLM 5.3 Flash didn’t. Both solved 8; neither solved 11.
stet-15439c21Passed testsstet-commit-5d766ddeEquivalent to the merged fixstet-43493c52Passed testsstet-commit-f8ae1177Passed testsstet-commit-342d0ac0Equivalent to the merged fixA 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 | GLM 5.3 Flash | GPT 6 Luna max | 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 | Failed | GLM 5.3 Flash only |
stet-66762b4c | Passed tests | Equivalent to the merged fix | Both |
stet-commit-f0caded9 | Passed tests | Equivalent to the merged fix | Both |
stet-commit-5d766dde | Failed | Equivalent to the merged fix | GPT 6 Luna max only |
stet-43493c52 | Failed | Passed tests | GPT 6 Luna max only |
stet-6f84e978 | Failed | Failed | Neither |
stet-commit-f8ae1177 | Failed | Passed tests | GPT 6 Luna max only |
stet-2450ca2d | Failed | Failed | Neither |
stet-commit-342d0ac0 | Failed | Equivalent to the merged fix | GPT 6 Luna max only |
stet-commit-60a0e6c2 | Failed | Failed | Neither |
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 | GLM 5.3 Flash | GPT 6 Luna max |
|---|---|---|
| Rank | 13 of 14 | 8 of 14 |
| Stet score | 31.095% range 21.5–41.5 | 50.095% range 40.1–59.3 |
| Solved | 9/249 passed, 0 judged equivalent | 12/248 passed, 4 judged equivalent |
| Passed review | 6/231 not recorded | 6/24 |
| Matched merged change | 6/24 | 11/24 |
| Cost / task | $0.41 | $0.37 |
| How cost was recorded | estimated from tokens | reported by its harness |
| Agent time / task | 26.7 min | 59.7 min |
| Ran in | Daytona sandbox | Daytona sandbox |
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.
GPT 6 Luna max matched the merged change more often: 7 of 11 to GLM 5.3 Flash’s 6.
| GLM 5.3 Flash | GPT 6 Luna max | |
|---|---|---|
| Matched merged change | 6/11 | 7/11 |
| Passed review | 6/101 not recorded | 6/11 |
| Cost / task | $0.06 | $0.09 |
| Agent time / task | 10.2 min | 19.0 min |
7 tasks
7 tasks whose instructions start with “Fix”. Their merged changes were 4–29 lines in 1–2 files.
Both matched the merged change on 5 of 7.
GLM 5.3 Flash passed review more often: 5 of 7 to GPT 6 Luna max’s 4.
| GLM 5.3 Flash | GPT 6 Luna max | |
|---|---|---|
| Matched merged change | 5/7 | 5/7 |
| Passed review | 5/7 | 4/7 |
| Cost / task | $0.05 | $0.08 |
| Agent time / task | 9.4 min | 17.1 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 Luna max matched the merged change more often: 4 of 13 to GLM 5.3 Flash’s 0.
| GLM 5.3 Flash | GPT 6 Luna max | |
|---|---|---|
| Matched merged change | 0/13 | 4/13 |
| Passed review | 0/13 | 0/13 |
| Cost / task | $0.70 | $0.61 |
| Agent time / task | 40.7 min | 94.2 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.