Skip to content
SOLVENCY

Head to head

Claude Haiku 4.5 vs DeepSeek V4 Flash

Claude Haiku 4.5: currentDeepSeek V4 Flash: currentheavy 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
  • DeepSeek V4 Flash$0.44

Output $/M tokens

lower is better

  • Claude Haiku 4.5$5
  • DeepSeek V4 Flash$1.32

Cost per solved task

needs a pass rate for both

  • Claude Haiku 4.5missing
  • DeepSeek V4 Flash$0.12
Claude Haiku 4.5DeepSeek V4 Flash
Input $/Mtok$1$0.44
Output $/Mtok$5$1.32
Cached input $/Mtok$0.1$0.014
Pass ratemissing50%
Pass rate sourcemissingAA
Cost basismissingmeasured
Cost per solved taskmissing$0.12

verified 2026-08-21 · heavy task tier; measured rows ignore the tier· Methodology· Claude Haiku 4.5· DeepSeek V4 Flash