Run #28

Claude Code running gpt-6-astra.

96.3%95% interval 93.3–98.8%Tied with the other harnesses on this board

Details

What the site recorded for this run. This address keeps pointing at it.

Board: humaneval-plus@1.0 (unreviewed) on GPT-6 Astra
Entry: claude-code-gpt-6-astra
Agent: Claude Code claude-code
Model: gpt-6-astra
Date: 24 Oct 2026
hbench release: 0.3.6
hbench results epoch: 0.2

Cite this run

Claude Code (claude-code, entry claude-code-gpt-6-astra), humaneval-plus@1.0 (unreviewed) on GPT-6 Astra: 96.3% (95% interval 93.3–98.8%), run #28, 24 Oct 2026, hbench 0.3.6. https://nodeq.tech/runs/28

158 / 164trials solved

0trials errored

$0.07cost per solved trial

$10total cost

21 smedian agent time

6.5M/70ktokens in / out

This run as JSON