SlopCodeBench 10/2026 – Opus 5.5, Astra wins overall, xhigh considered harmful
github.comdhorthy
SlopCodeBench 10/2026 evaluates coding models against progressively disclosed feature lists without advance knowledge, testing how well models handle evolving requirements. Opus 5.5 leads overall performance, while Astra wins in specific categories, with results suggesting xhigh complexity settings produce harmful outcomes.
Why it mattersDevelopers choosing coding AI models need benchmarks that simulate real-world scenarios where requirements change mid-project; this benchmark's progressive disclosure approach better reflects actual development workflows than static test suites.
From the source: Quick update here since a lot of folks asked about what slopcodebench is / where it fits in Slop code bench is unique among benchmarks in that it makes a model work against a progressively disclosed list of features (checkpoints) that it doesn’t know it advance, inheriting the same codebase…Read at github.com