Claude Haiku 5.5 (Medium) vs Claude Opus 5.5 (Max, Default Fallback)

What the captured data shows

Claude Haiku 5.5 Medium is listed at lower token rates than Claude Opus 5.5. Opus has the higher intelligence index, while Haiku has a shorter reported first-token delay and higher output throughput in the captured records. These are aggregate and operational measurements rather than evidence for a particular professional task.

Which model fits your workflow

For an existing Opus workflow, select examples where a mistake would be visible and test Haiku Medium against the same requirements. Track revisions and retries. Retain the settings in the record so a later change in configuration is not confused with a model improvement.

Decision takeaway

The snapshot describes a lower-cost, faster-starting Haiku configuration alongside a higher-index Opus entry. Task acceptance should decide the practical choice.

These figures come from Franklin’s Artificial Analysis-sourced model records captured on 2026-10-11. Read the individual model entries for the data snapshot: Claude Haiku 5.5 (Medium) and Claude Opus 5.5 (Max, Default Fallback). Missing measurements remain unavailable; this page does not report a new hands-on test.