Skip to the comparison

GLM 5.3 Flash high vs SWE-2 max

GLM 5.3 Flash high leads on speed and cost; no clear quality difference. Different harnesses: GLM 5.3 Flash high ran codex, SWE-2 max ran devin. This compares setups, not models alone.

Quality
Even
No clear difference
1–6 head to head, 13 tied, 4 undecided
Speed
2.6× faster
GLM 5.3 Flash high
26.6 min vs 68.3 min per task; 26.6 min from 20/24 tasks
Cost
6.2× cheaper
GLM 5.3 Flash high
$0.45 vs $2.76 per task; $0.45 from 20/24 tasks; $2.76 from 23/24 tasks

What this comparison is

  • 1 real repositoryStet-AI/stet, private; not public benchmark scores
  • 24 merged changeseach run end to end by the coding agent from the repo as it was before the change
  • Graded against the merged changeby the repo’s own tests and by judges comparing each patch with it
  • Quality, cost and timea smaller sample than public benchmarks, but real agent work on a real codebase

24 tasksData from

Stet score

GLM 5.3 Flash highcodex · high reasoningRank 10 of 18

49.095% range 40.9–57.2

SWE-2 maxdevin · max reasoningRank 1 of 18

62.395% range 53.4–71.6

Their 95% score ranges overlap.

Quality

Head-to-head matchups

GLM 5.3 Flash high won1

Tied13

4 undecided

SWE-2 max won6

SWE-2 max’s win share 62%, 95% range 50–74%Ties are within judge noise.

Other quality measures

Other quality measures
MetricGLM 5.3 Flash highSWE-2 max
Code review pass rate33%18 tasks33%24 tasks
Checklist score0.7220 tasks0.8424 tasks
Equivalence rate30%20 tasks54%24 tasks

These measures are not part of the headline Stet score and carry no confidence interval, so a small gap may be noise. The marked side has the higher recorded value, not a proven lead.

Task by task

GLM 5.3 Flash high solved 1 task that SWE-2 max didn’t, and SWE-2 max solved 4 that GLM 5.3 Flash high didn’t. Both solved 9; neither solved 6.

On the 9 tasks both solved, judges preferred SWE-2 max’s patch on 1 and called 8 even.

1Only GLM 5.3 Flash high solved

  • 1,312 changed lines in 9 filesstet-commit-f8ae1177Judges preferred GLM 5.3 Flash high’s patch

4Only SWE-2 max solved

  • 481 changed lines in 3 filesstet-commit-5e075641Judges preferred SWE-2 max’s patch
  • 552 changed lines in 11 filesstet-commit-60a0e6c2Judges preferred SWE-2 max’s patch
  • 564 changed lines in 10 filesstet-commit-a8716b88Judges preferred SWE-2 max’s patch
  • 640 changed lines in 8 filesstet-commit-342d0ac0Judges preferred SWE-2 max’s patch

9Both solved

  • 5 changed lines in 2 filesstet-15439c21
  • 6 changed lines in 1 filestet-3169aefa
  • 9 changed lines in 1 filestet-89dfbc27
  • 10 changed lines in 1 filestet-232b50ba
  • 23 changed lines in 2 filesstet-5276aa3f
  • 29 changed lines in 1 filestet-b28f4fe7
  • 84 changed lines in 1 filestet-66762b4c
  • 113 changed lines in 4 filesstet-commit-f0caded9
  • 941 changed lines in 7 filesstet-commit-bbbbae09Judges preferred SWE-2 max’s patch

6Neither solved

  • 4 changed lines in 1 filestet-43493c52
  • 149 changed lines in 2 filesstet-commit-e928166fJudges preferred SWE-2 max’s patch
  • 763 changed lines in 8 filesstet-commit-87814583
  • 977 changed lines in 5 filesstet-commit-e41c2e2d
  • 1,236 changed lines in 16 filesstet-commit-d2ab0c25
  • 1,420 changed lines in 19 filesstet-commit-2826765e

Bar: changed lines in the merged change, log scale

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. 4 of the 24 tasks are left out because one of the two has no scored result on them.

All 20 compared tasks
Task by task
TaskGLM 5.3 Flash highSWE-2 maxSolved byBetter patch
stet-89dfbc27 · 9 changed lines in 1 filePassed tests$0.04 · 6.6 minPassed tests$0.71 · 4.5 minBothEven
stet-b28f4fe7 · 29 changed lines in 1 filePassed tests$0.04 · 7.2 minPassed tests$2.02 · 16.7 minBothEven
stet-232b50ba · 10 changed lines in 1 filePassed tests$0.04 · 10.7 minPassed tests$1.56 · 11.9 minBothEven
stet-3169aefa · 6 changed lines in 1 filePassed tests$0.05 · 7.0 minPassed tests$1.03 · 5.3 minBothEven
stet-5276aa3f · 23 changed lines in 2 filesPassed tests$0.08 · 12.7 minPassed tests$1.03 · 10.7 minBothEven
stet-66762b4c · 84 changed lines in 1 filePassed tests$0.06 · 12.6 minPassed tests$1.77 · 11.2 minBothEven
stet-commit-bbbbae09 · 941 changed lines in 7 filesPassed tests$0.36 · 30.6 minPassed tests$0.61 · 89.7 minBothSWE-2 max
stet-43493c52 · 4 changed lines in 1 fileFailed$0.07 · 14.0 minFailed$3.52 · 35.3 minNeitherEven
stet-commit-f0caded9 · 113 changed lines in 4 filesPassed tests$0.45 · 20.9 minPassed tests$3.91 · 22.6 minBothEven
stet-15439c21 · 5 changed lines in 2 filesPassed tests$0.05 · 11.1 minPassed tests$2.24 · 34.3 minBothEven
stet-commit-f8ae1177 · 1,312 changed lines in 9 filesPassed tests$0.58 · 40.1 minFailed300.0 minGLM 5.3 Flash high onlyGLM 5.3 Flash high
stet-commit-342d0ac0 · 640 changed lines in 8 filesFailed$0.30 · 37.6 minEquivalent to the merged fix$2.39 · 146.4 minSWE-2 max onlySWE-2 max
stet-commit-5e075641 · 481 changed lines in 3 filesFailed$0.32 · 36.2 minPassed tests$2.64 · 86.0 minSWE-2 max onlySWE-2 max
stet-commit-60a0e6c2 · 552 changed lines in 11 filesFailed$0.27 · 22.2 minPassed tests$0.66 · 106.1 minSWE-2 max onlySWE-2 max
stet-commit-2826765e · 1,420 changed lines in 19 filesFailed$0.36 · 31.5 minFailed$7.00 · 144.5 minNeitherEven
stet-commit-a8716b88 · 564 changed lines in 10 filesFailed$0.40 · 32.7 minEquivalent to the merged fix$1.29 · 102.3 minSWE-2 max onlySWE-2 max
stet-commit-e928166f · 149 changed lines in 2 filesFailed$0.17 · 13.2 minFailed$1.49 · 59.4 minNeitherSWE-2 max
stet-commit-87814583 · 763 changed lines in 8 filesFailed$1.75 · 57.0 minFailed$5.19 · 73.6 minNeitherEven
stet-commit-d2ab0c25 · 1,236 changed lines in 16 filesFailed$0.61 · 34.5 minFailed$5.66 · 68.9 minNeitherEven
stet-commit-e41c2e2d · 977 changed lines in 5 filesFailed$2.92 · 92.9 minFailed$6.08 · 69.8 minNeitherEven

Cost and time on each task

Each dot is one task, on a log scale: filled where the setup solved it, hollow where it didn’t. The line marks the median.

Cost per task

GLM 5.3 Flash highmean $0.45 · median $0.29

SWE-2 maxmean $2.67 · median $2.02from 19/20 tasks

GLM 5.3 Flash high cost less on all 19 tasks.

Agent time per task

GLM 5.3 Flash highmean 26.6 min · median 21.5 min

SWE-2 maxmean 70.0 min · median 64.2 min

GLM 5.3 Flash high finished faster on 15 of 20 tasks; SWE-2 max on 5.

Where each setup is the better choice

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.

Small fixes

Cheaper here without solving fewer: GLM 5.3 Flash high

10 tasks · 2–84 changed lines in 1–2 files

GLM 5.3 Flash high solved as many, 7 of 8, at 32× lower cost per task.

GLM 5.3 Flash high was 1.6× faster.

Small fixes
GLM 5.3 Flash highSWE-2 max
Solved7/87/8
Cost / task$0.05$1.73
Agent time / task10.2 min16.2 min

Large, cross-cutting changes

14 tasks · 113–1,420 changed lines in up to 19 files

SWE-2 max solved 6 of 12 to GLM 5.3 Flash high’s 3.

GLM 5.3 Flash high was 2.8× faster.

Large, cross-cutting changes
GLM 5.3 Flash highSWE-2 max
Solved3/126/12
Cost / task$0.71$3.36from 11/12 tasks
Agent time / task37.4 min105.8 min

Side by side

Side by side
MetricGLM 5.3 Flash highSWE-2 max
Harnesscodexdevin
Reasoning efforthighmax
Rank10 of 181 of 18
Stet score49.095% range 40.9–57.262.395% range 53.4–71.6
Solved10/2410 passed, 0 judged equivalent16/2413 passed, 3 judged equivalent
Cost / task$0.45from 20/24 tasks$2.76from 23/24 tasks
Agent time / task26.6 minfrom 20/24 tasks68.3 min

Questions

Which is better for coding, GLM 5.3 Flash high or SWE-2 max?
No clear quality difference: SWE-2 max won 6, GLM 5.3 Flash high won 1, 13 tied, 4 undecided; GLM 5.3 Flash high leads on cost and agent time: 6.2× cheaper and 2.6× faster. Measured on 24 merged changes from Stet-AI/stet.
How were GLM 5.3 Flash high and SWE-2 max tested?
Both coding agents worked the same 24 merged changes from the private Stet-AI/stet repository, each in the real repo as it was before the change. The repo’s tests and judges graded every patch against the merged change; no public benchmark scores are used. Stet score: A setup’s average chance of beating another setup on a task, from 0 to 100; 50 is even with the field.

See which setup wins on your repo

Stet replays your merged changes, scores quality above pass/fail, and delivers comparison reports through your agent.

Get started