agent-cost-optimizer / eval /swe_bench_results.json
narcolepticchicken's picture
Upload eval/swe_bench_results.json with huggingface_hub
2462002 verified
{
"always_frontier": {
"success": 0.782,
"avg_cost": 0.3166872804999999,
"n": 500
},
"always_cheap": {
"success": 0.632,
"avg_cost": 0.014239461958399993,
"n": 500
},
"aco_v8": {
"success": 0.756,
"avg_cost": 0.29145885426799994,
"n": 500
},
"oracle": {
"success": 0.87,
"avg_cost": 0.05456037742200003,
"n": 500
},
"aco_v9_feedback": {
"success": 0.826,
"avg_cost": 0.4847524783679998,
"n": 500
}
}