List your company

Task Planning on EB-ALFRED Base

96Success Rate (SR)Claude-3.5-Sonnet
25.2843.646280.36Aug 2, 2025Sep 8, 2025Oct 15, 2025Nov 21, 2025Dec 28, 2025Feb 3, 2026Mar 12, 2026

Evaluation results

MethodLinks
Claude-3.5-SonnetType=Closed-source2025.089697.5
Llama-3.2-90B-Vision-InsType=Open-source2025.089494.5
Gemini-1.5-ProType=Closed-source2025.089292.5
Claude-3.7-SonnetType=Closed-source2025.0890—
GPT-4oType=Closed-source2025.088690.7
RoboMemoryType=Ours, Backbone=Qwen2.5-VL-72B-Ins2025.088688
InternVL3-78BType=Open-source2025.0884—
Gemini-2.0-flashType=Closed-source2025.088282
InternVL2.5-78BType=Open-source2025.088082
ReflexionType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.088084.2
VoyagerType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.087687
GPT-4o-miniType=Closed-source2025.087477.5
Qwen2.5-VL-72B-InsType=Open-source2025.0874—
Claude-3.5-SonnetType=Closed-source2025.087272
BrainMemEvaluation Model=Qwen2.5-VL-72B-Ins2026.037278.2
Gemini-1.5-ProType=Closed-source2025.087074.3
Claude-3.7-SonnetType=Closed-source2025.0868—
RoboMemoryType=Ours, Backbone=Qwen2.5-VL-72B-Ins2025.086875.5
RoboMemoryEvaluation Model=Qwen2.5-VL-72B-Ins2026.036875.5
GPT-4oType=Closed-source2025.086474
Gemini-2.0-flashType=Closed-source2025.086265.7
CradleType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.086267
InternVL2.5-38BType=Open-source2025.086061.5
VoyagerType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.085673.2
VoyagerEvaluation Model=Qwen2.5-VL-72B-Ins2026.035673.2
CradleType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.085467.9
CradleEvaluation Model=Qwen2.5-VL-72B-Ins2026.035467.9
Qwen2.5-VL-72B-InsType=Open-source2025.0850—
ReflexionType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.084854
ReflexionEvaluation Model=Qwen2.5-VL-72B-Ins2026.034854
Llama-3.2-90B-Vision-InsType=Open-source2025.083843.7
InternVL2.5-78BType=Open-source2025.083842.3
InternVL3-78BType=Open-source2025.0838—
RoboOSType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.083847.8
InternVL2.5-38BType=Open-source2025.083637.3
GPT-4o-miniType=Closed-source2025.083447.8
RoboOSType=Baselines, Backbone=Qwen2.5-VL-72B-Ins2025.083238.4
RoboOSType=Baselines, Backbone=RoboBrain2-32B2025.083237.2
RoboOSEvaluation Model=Qwen2.5-VL-72B-Ins2026.033238.4
RoboOSType=Baselines, Backbone=RoboBrain2-32B2025.082834.8