Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embodied Task Planning on ALFWorld textual observation (unseen)
Loading...
94
Success Rate
RoboAgent
71.536
77.368
83.2
89.032
Apr 9, 2026
Success Rate
Updated 9d ago
Evaluation Results
Method
Method
Links
Success Rate
RoboAgent
Base Model=Qwen2.5-VL-3B
2026.04
94
BPO
Base Model=Llama3.1-8B
2026.04
89.6
DynaMind
Base Model=Qwen2.5-7B
2026.04
89.1
SEEA-R1
Base Model=Qwen2.5-7B
2026.04
85.1
LAC
Base Model=Gemma-7B
2026.04
84
Zero-Shot
Base Model=GPT-4o
2026.04
83.6
MPO
Base Model=Llama3.1-8B
2026.04
79.1
IPR
Base Model=Llama-2-7B
2026.04
74.7
ETO
Base Model=Llama2-7B
2026.04
72.4
Feedback
Search any
task
Search any
task
