Coding reasoning comparison · Figure

Astra, Sol and Fable at selected reasoning levels

GPT-6 Astra · GPT-5.6 Sol · Claude Fable 5.1 · Artificial Analysis · 10 Sep 2026 · Read the full comparison

Bars are coloured by model. The dashed line marks Astra medium, the baseline for each comparison. Hover or focus a bar for its ratio to that baseline.

Coding: benchmark pass rates. Tokens/cost: mixed Intelligence Index v4.3 tasks. Latency: separate API prompts; first answer, not a completed patch.