GPT 6 Sol medium leads on quality and agent time: 26.6 points higher and 1.7× faster; GLM 5.3 Flash leads on cost: 7.3× cheaper.
Stet score
GLM 5.3 FlashRank 13 of 14
31.095% range 21.5–41.5
GPT 6 Sol mediumRank 7 of 14
57.695% range 49.5–65.726.6 points higher
GLM 5.3 Flash solved 0 tasks that GPT 6 Sol medium didn’t, and GPT 6 Sol medium solved 5 that GLM 5.3 Flash didn’t. Both solved 9; neither solved 10.
stet-commit-5d766ddeEquivalent to the merged fixstet-43493c52Passed testsstet-6f84e978Passed 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 Sol medium | 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 | Passed tests | Equivalent to the merged fix | Both |
stet-commit-5d766dde | Failed | Equivalent to the merged fix | GPT 6 Sol medium only |
stet-43493c52 | Failed | Passed tests | GPT 6 Sol medium only |
stet-6f84e978 | Failed | Passed tests | GPT 6 Sol medium only |
stet-commit-f8ae1177 | Failed | Passed tests | GPT 6 Sol medium only |
stet-2450ca2d | Failed | Failed | Neither |
stet-commit-342d0ac0 | Failed | Equivalent to the merged fix | GPT 6 Sol medium 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 Sol medium |
|---|---|---|
| Rank | 13 of 14 | 7 of 14 |
| Stet score | 31.095% range 21.5–41.5 | 57.695% range 49.5–65.7 |
| Solved | 9/249 passed, 0 judged equivalent | 14/2411 passed, 3 judged equivalent |
| Passed review | 6/231 not recorded | 8/24 |
| Matched merged change | 6/24 | 10/24 |
| Cost / task | $0.41 | $3.01 |
| How cost was recorded | estimated from tokens | estimated from tokens |
| Agent time / task | 26.7 min | 15.4 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.
Both matched the merged change on 6 of 11.
| GLM 5.3 Flash | GPT 6 Sol medium | |
|---|---|---|
| Matched merged change | 6/11 | 6/11 |
| Passed review | 6/101 not recorded | 6/11 |
| Cost / task | $0.06 | $0.75 |
| Agent time / task | 10.2 min | 6.6 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 | GPT 6 Sol medium | |
|---|---|---|
| Matched merged change | 5/7 | 5/7 |
| Passed review | 5/7 | 5/7 |
| Cost / task | $0.05 | $0.60 |
| Agent time / task | 9.4 min | 5.4 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 GLM 5.3 Flash’s 0.
| GLM 5.3 Flash | GPT 6 Sol medium | |
|---|---|---|
| Matched merged change | 0/13 | 4/13 |
| Passed review | 0/13 | 2/13 |
| Cost / task | $0.70 | $4.92 |
| Agent time / task | 40.7 min | 22.9 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.