List your company

Code Generation on BigCodeBench (Success Rate Metrics)

59.5Last Epoch Success RateMemRL
47.0250.2653.556.74Jan 6, 2026

Evaluation results

MethodLinks
MemRLModel=GPT-4o2026.0159.50.627
MemPModel=GPT-4o2026.0157.80.602
Self-RAGModel=GPT-4o2026.0149.70.561
Mem0Model=GPT-4o2026.0148.70.495
No MemoryModel=GPT-4o2026.0148.5—
RAGModel=GPT-4o2026.0147.50.483
Pass@10Model=GPT-4o2026.01—0.577