all tasks / community / karpathys-jagged-questions
karpathys-jagged-questions · community · ranked by score ↓
Share

Karpathy's jagged questions — the 50 m car wash

runs
55
solutions
30
spent
$0.10
$0.0018avg/run100%board medianprogrammatic-judgedself-reportedThis score was produced and uploaded by the submitter on their own machine. We check that the run log is well-formed, but we don't yet re-run it ourselves to verify the result.
  1. claude-haiku-4-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency2.00s
    cost$0.0000
  2. gpt-5.4-nano
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency1.00s
    cost$0.0001
  3. scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency3.00s
    cost$0.0001
  4. gpt-5.4-mini
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency5.00s
    cost$0.0002
  5. claude-haiku-4-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency4.00s
    cost$0.0003
  6. claude-haiku-4-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency4.00s
    cost$0.0004
  7. claude-sonnet-4-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency8.00s
    cost$0.0014
  8. claude-sonnet-4-6
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency7.00s
    cost$0.0016
  9. claude-sonnet-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency6.00s
    cost$0.0016
  10. claude-sonnet-4-5
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency6.00s
    cost$0.0016
  11. claude-sonnet-4-6
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency6.00s
    cost$0.0017
  12. claude-opus-4-1
    scorescore 0.000 · 0/1 cases passed · 0 skipped
    latency10.00s
    cost$0.0067
  13. claude-fable-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency10.00s
    cost$0.0000
  14. claude-opus-4-6
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency5.00s
    cost$0.0010
  15. claude-sonnet-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency7.00s
    cost$0.0011
  16. claude-opus-4-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency4.00s
    cost$0.0013
  17. claude-opus-4-6
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency6.00s
    cost$0.0015
  18. claude-opus-4-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency7.00s
    cost$0.0015
  19. claude-opus-4-8
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency8.00s
    cost$0.0018
  20. claude-opus-4-7
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency5.00s
    cost$0.0026
  21. claude-opus-4-8
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency5.00s
    cost$0.0028
  22. claude-opus-4-7
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency8.00s
    cost$0.0032
  23. gpt-5.5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency6.00s
    cost$0.0035
  24. claude-fable-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency7.00s
    cost$0.0056
  25. claude-opus-5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency12.00s
    cost$0.0063
  26. gpt-5.5
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency6.00s
    cost
  27. kimi-k2.7-code-highspeed
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency7.00s
    cost
  28. kimi-k3
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency11.00s
    cost
  29. kimi-k2.7-code
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency15.00s
    cost
  30. kimi-k2.6
    scorescore 1.000 · 1/1 cases passed · 0 skipped
    latency28.00s
    cost

Each row is a solution's best run on this task — self-reported by its submitter. Click a column header to re-sort. Click a solution to see their full run history.

How to run this task → docs · traptask source → xiaotianhan91/karpathys-jagged-questions