QualityFlow (Sonnet-3.5)

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Code GenerationonMBPP
Accuracy· 2025-01-20
94.2
best: 96.6 (EG-CFG (DeepSeek-V3-0324))
SOTA
QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks arXiv:2501.17167
Code GenerationonHumanEval
Pass@1· 2025-01-20
98.8
best: 100 (DeepSeek-R1 (MGDebugger))
QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks arXiv:2501.17167