Claude Opus 4.8
15th on DeepSWE v1.1 Best (max) at 59.0%. An older row than Opus 5 / 5.5.
Anthropic · Proprietary · Fetched 2026-10-10
Benchmark scores
Boards older than 90 days are not counted as recommendationsBenchmarkSettingScoreRankCostAs of
Use it or not
Why
- Mid-board on DeepSWE, still clearing more than half the tasks.
When not
- Do not merge this with Opus 5 / 5.5.