cmp-probe · status: completed

Bench results

capability fixture model status score cost($) dur(s)
eval uk-1 aisubscription error - 0.000000 600.5
eval uk-1 openrouter/free error - 0.000000 600.5
eval uk-1 anthropic/claude-sonnet-4.6 error - 0.000000 12.1
eval uk-1 moonshotai/kimi-k2.6 error - 0.000000 21.1

Winners

  • score: aisubscription
  • cost: aisubscription
  • speed: anthropic/claude-sonnet-4.6
  • overall: anthropic/claude-sonnet-4.6