List your company

Machine Learning Engineering on MLE-Bench full official

80.3Medal Rate (Low)MLEvolve
1.15621.70342.2562.797Mar 2, 2026Mar 17, 2026Apr 2, 2026Apr 18, 2026May 3, 2026May 19, 2026Jun 4, 2026

Evaluation results

MethodLinks
MLEvolveTime (h)=12, Backbone=Gemini-3.1-Pro-preview, Code Availability=Open-Source2026.0680.36446.765.31007634.7
MARS+Time (h)=24, Backbone=Gemini-3-Pro-preview, Code Availability=Proprietary2026.0678.860.544.462.710074.233.8
AIBuildAITime (h)=24, Backbone=Claude-Opus-4.6, Code Availability=Proprietary2026.0677.361.446.763.110071.125.8
ML-Master 2.0Time (h)=24, Backbone=DeepSeek-V3.2-Speciale, Code Availability=Open-Source2026.0675.850.942.256.495.663.119.6
MARSTime (h)=24, Backbone=Gemini-3-Pro-preview, Code Availability=Proprietary2026.0674.252.637.85698.765.831.1
Gome (GPT-5)Time=12h, GPU=V1002026.0368.221.122.235.19645.316.4
MLE-STAR-Pro-1.5Time (h)=24, Backbone=Gemini-2.5-Pro, Code Availability=Proprietary2026.0668.234.233.34493.852.919.1
R&D-AgentTime (h)=12, Backbone=gpt-5, Code Availability=Open-Source2026.0668.221.122.235.153.340.416.4
LeerooTime (h)=24, Backbone=Gemini-3-Pro-preview, Code Availability=Open-Source2026.0668.244.74050.750.750.721.3
FM-AgentTime (h)=24, Backbone=Gemini-2.5-Pro, Code Availability=Proprietary2026.0662.136.833.343.696.951.622.7
Gome (o3)Time=12h, GPU=V1002026.0359.120.222.232.594.243.117.3
AIRA (o3)Time=24h, GPU=H2002026.03552221.731.697.545.517.3
AIRA-DojoTime (h)=24, Backbone=o3, Code Availability=Open-Source2026.06552221.731.697.545.517.3
ML-Master (GPT-5)Time=12h, GPU=V1002026.0351.510.517.82498.234.712.4
Gome (DeepSeek-R1)Time=12h, GPU=V1002026.03509.72023.488.935.112.4
ML-MasterTime (h)=12, Backbone=DeepSeek-R1, Code Availability=Open-Source2026.0648.520.224.429.393.344.917.3
ML-Master (DeepSeek-R1)Time=12h, GPU=V1002026.03479.72022.791.632.912.4
ML-Master (o3)Time=12h, GPU=V1002026.0342.412.32022.794.231.513.3
AIDETime (h)=24, Backbone=o1-preview, Code Availability=Open-Source2026.0635.98.511.717.182.829.49.4
AIDE (o1-preview)Time=24h, GPU=A10†2026.0334.38.81016.982.829.49.4
AIDE (gpt-4o-24-08)Time=24h, GPU=A10†2026.03193.25.68.654.914.45
OpenHands (gpt-4o-24-08)Time=24h, GPU=A10†2026.0311.52.21.95.1527.12.7
MLAB (gpt-4o-24-08)Time=24h, GPU=A10†2026.034.2001.344.31.90.8