humaneval-plus@1.0 (unreviewed) on GPT-5.6 Luna
164 tasks, 1 attempt, open network. Updated 24 Oct 2026.
164 tasks from humaneval-plus@1.0 on ghcr.io. Nobody has reviewed this dataset, so a score here says nothing about the benchmark it's named after.
Compare
Pick two harnesses to compare them task by task.
Codex 0.164.0 against Claude Code 2.1.309, task by task
Over the 164 tasks both ran, 1 attempt each
Codex 0.164.0 minus Claude Code 2.1.309, pass rate
+1.8 points
95% interval −0.6 to +4.9
The interval includes zero, so the paired test can't tell the two apart.
Where the tasks landed
- 4Only Codex 0.164.0 solved
- 153Both solved
- 1Only Claude Code 2.1.309 solved
- 6Neither solved
Of 164 tasks, 4 only Codex 0.164.0 solved, 1 only Claude Code 2.1.309 solved, 153 both solved and 6 neither did.
| Task | Reward | Tokens | Time | |||
|---|---|---|---|---|---|---|
| Codex 0.164.0 | Claude Code 2.1.309 | Codex 0.164.0 | Claude Code 2.1.309 | Codex 0.164.0 | Claude Code 2.1.309 | |
humaneval-plus-137 | 1 | 0 | 46k | 136k | 17 s | 1.4 min |
humaneval-plus-151 | 1 | 0 | 45k | 42k | 14 s | 21 s |
humaneval-plus-22 | 1 | 0 | 33k | 31k | 11 s | 15 s |
humaneval-plus-76 | 1 | 0 | 73k | 33k | 28 s | 44 s |
humaneval-plus-83 | 0 | 1 | 56k | 41k | 19 s | 21 s |