List your company

Systems Optimization on LLM-SQL

0.7331Final ScoreSwarmResearch
0.6903560.7014530.712550.723647Apr 2, 2026Apr 17, 2026May 2, 2026May 17, 2026Jun 1, 2026Jun 16, 2026Jul 2, 2026

Evaluation results

MethodLinks
SwarmResearchagent_framework=Claude Code, backbone=Opus 4.6, budget_per_task=$502026.070.7331——
CORALAgent Model=Claude Code + Opus 4.62026.040.73153.315
SOTA2026.070.731——
SOTA2026.040.73——
EvoXAgent Model=Claude Code + Opus 4.62026.040.726683
EvoXiterations=100, backbone=Opus 4.62026.070.7253——
ShinkaEvolveAgent Model=Claude Code + Opus 4.62026.040.72423.821
SEAEVOSHINKABackbone=Gemini-3-Flash2026.040.72——
CORALagent_framework=Claude Code, backbone=Opus 4.6, budget_per_task=$502026.070.7195——
OpenEvolveAgent Model=Claude Code + Opus 4.62026.040.7166.7100
GEPABackbone=Gemini-3-Flash2026.040.7158——
OpenEvolveBackbone=Gemini-3-Flash2026.040.7117——
SEAEVOOPENBackbone=Gemini-3-Flash2026.040.7104——
ShinkaEvolveBackbone=Gemini-3-Flash2026.040.6998——
Human/SOTA2026.040.692——