Skip to content
Hbench
hbench

humaneval-plus@1.0 (unreviewed) on codex-auto-review

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.

Build a run for this board

Value

What a task cost each configuration. Claude Code used 22% more tokens than Codex, and 93% more time per task.

$ per solved task

Lower is cheaper. Dollars per task divided by the pass rate.

Embed this chart

Harness$ per solved task
Claude Code2.1.309$0.10$0.10 a task
Not reported: Codex

Tokens per task

Lower is leaner. Input plus output, cached input included.

Embed this chart

HarnessTokens per task
Codex0.164.0 · Reference48k×1.00
Claude Code2.1.30958k×1.22

Agent time per task

Lower is faster. Agent start to finish; setup and tests left out.

Embed this chart

HarnessTime per task
Codex0.164.0 · Reference18 s×1.00
Claude Code2.1.30934 s×1.93

Pass rate against tokens per task

Up and to the left is better. Not offered: dollars, which fewer than two rows report. Pass-rate axis starts at 85%.

Embed this chart

85%90%95%100%45k50k55k60kTokens per task, input and outputMore tokens for a lower pass rate than CodexCodex0.164.0 · 48k · 94.5%Claude Code2.1.309 · 58k · 93.3%
  • On the frontier: no other row has a higher pass rate for less tokens
  • Off the frontier
  • 95% interval of the pass rate
On the frontier, least tokens first: Codex 0.164.0, 48k tokens a task, 94.5%.
  • 2 runs
  • Updated 24 Oct 2026
  • Method