Skip to content
SOLVENCY

Head to head

Claude Haiku 4.5 vs GPT-5

Claude Haiku 4.5: currentGPT-5: legacyheavy 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

  • Claude Haiku 4.5$1
  • GPT-5$1.25

Output $/M tokens

lower is better

  • Claude Haiku 4.5$5
  • GPT-5$10

Cost per solved task

needs a pass rate for both

  • Claude Haiku 4.5missing
  • GPT-5$1.66
Claude Haiku 4.5GPT-5
Input $/Mtok$1$1.25
Output $/Mtok$5$10
Cached input $/Mtok$0.1$0.125
Pass ratemissing88%
Pass rate sourcemissingAider
Cost basismissingmodelled
Cost per solved taskmissing$1.66

verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· Claude Haiku 4.5· GPT-5