the only figure that people who use claude code and codex care about if their workload mimics deepswe: more and cheaper intelligence from maxed gpt 5.5 than maxed opus 4.8 intelligence of opus 4.8 max and opus 4.8 xhigh is similar
Claude Opus 4.8 has landed on DeepSWE Bench, posting a 58% Pass@1 and taking #2 overall behind GPT-5.5. It continues a broader trend: slightly behind on raw score, but among the most reliable and efficient coding models across recent benchmarks.
