A 27B dense model beating the 397B-A17B MoE on coding benchmarks is quite cool, and a better signal than the open-source-catches-closed headline it'll get flattened into.
27B Dense Model Outperforms 397B MoE on Coding Benchmarks
By
–
By
–
A 27B dense model beating the 397B-A17B MoE on coding benchmarks is quite cool, and a better signal than the open-source-catches-closed headline it'll get flattened into.