Claude Haiku 5.5 (Medium) vs GPT-6 Astra (Max)

What the captured data shows

Claude Haiku 5.5 Medium has lower listed rates than GPT-6 Astra Max and a lower intelligence index. The recorded Haiku timing values also show a shorter first-token delay and higher output throughput. This pair compares a Medium configuration with a Max configuration, which readers need to know when interpreting the gap.

Which model fits your workflow

Test whether the task benefits from the additional reasoning represented by the Astra configuration. Use examples with explicit acceptance criteria and preserve failed attempts. For high-volume work, calculate cost per accepted result rather than assuming the lowest listed rate produces the cheapest successful workflow.

Decision takeaway

The dataset shows a lower-cost Haiku entry and a higher-index Astra entry. The exact task should determine whether the added cost is justified.

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 GPT-6 Astra (Max). Missing measurements remain unavailable; this page does not report a new hands-on test.