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

What the captured data shows

The Max entries for Claude Haiku 5.5 and GPT-6 Astra both have reported intelligence indexes. Astra’s index is higher, while Haiku’s listed token prices are substantially lower. Haiku also has higher reported output throughput here, but throughput alone does not describe the time required to finish a useful task.

Which model fits your workflow

Keep the Max setting on both sides when trying your own examples. Measure completion time and accepted output separately, especially for a job that can involve substantial reasoning before text appears. The dataset contains no individual benchmark results to identify a specific task advantage.

Decision takeaway

Astra leads the recorded aggregate index; Haiku offers lower listed rates. Verify the quality gap on the work you actually plan to run.

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