SWE-Serve: Benchmarking Agentic Engineering For Production Inference Serving
By making the production correctness gap directly measurable, SWE-Serve enables the field to track whether future agents move beyond completing tasks locally to achieving production correctness, and exposes a substantial gap between completing tasks locally and achieving production correctness.
Jennifer Williams, David P. Farris, Jeffrey J. Farris et al.
· 0 citations