[KongchangAI]
Benchmark

CodeMemoryBench

A benchmark for evaluating the capabilities of code memory systems, measuring the accuracy of code context recall and the rate at which software engineering problems are resolved directly.

Timeline (last 90 days)

Oct 19

On CodeMemoryBench, the Code Memory layer achieves 64.2% code context recall accuracy and a 35.6% direct resolution rate for SWE-level issues.

Unverified50%

All Facts (1)

Source Articles