Skip to content
Hbench
hbench

humaneval-plus@1.0 (unreviewed) on GPT-6 Sol

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. Codex used 36% more tokens than Claude Code, and 12% less time per task.

$ per solved task

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

Embed this chart

Harness$ per solved task
Codex0.164.0$0.03$0.03 a task
Claude Code2.1.309 · Reference$0.07$0.07 a task

Tokens per task

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

Embed this chart

HarnessTokens per task
Claude Code2.1.309 · Reference42k×1.00
Codex0.164.056k×1.36

Agent time per task

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

Embed this chart

HarnessTime per task
Codex0.164.022 s×0.88
Claude Code2.1.309 · Reference25 s×1.00

Pass rate against tokens per task

Up and to the left is better. Pass-rate axis starts at 90%.

Embed this chart

90%95%100%35k40k45k50k55k60kTokens per task, input and outputMore tokens for a lower pass rate than Claude CodeClaude Code2.1.309 · 42k · 95.1%Codex0.164.0 · 56k · 95.1%
  • 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: Claude Code 2.1.309, 42k tokens a task, 95.1%.
  • 2 runs
  • Updated 24 Oct 2026
  • Method