Skip to content
SOLVENCY

Head to head

GPT-5.6 Luna vs Claude Opus 4

GPT-5.6 Luna: currentClaude Opus 4: retiredheavy tier

No verdict

Cost per solved task needs a published pass rate for both models. At least one has none, so it is reported as missing rather than estimated.

verified 2026-08-21 · prices from each provider's pricing page

Input $/M tokens

lower is better

  • GPT-5.6 Luna$0.2
  • Claude Opus 4$15

Output $/M tokens

lower is better

  • GPT-5.6 Luna$1.2
  • Claude Opus 4$75

Cost per solved task

needs a pass rate for both

  • GPT-5.6 Lunamissing
  • Claude Opus 4$18.75
GPT-5.6 LunaClaude Opus 4
Input $/Mtok$0.2$15
Output $/Mtok$1.2$75
Cached input $/Mtok$0.02$1.5
Pass ratemissing72%
Pass rate sourcemissingAider
Cost basismissingmodelled
Cost per solved taskmissing$18.75

verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· GPT-5.6 Luna· Claude Opus 4