Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Text-based agent interaction on TextWorld Quest
Loading...
88
Accuracy
Agent-BRACE
59.92
67.21
74.5
81.79
May 12, 2026
Accuracy
Steps Taken
Updated 21d ago
Evaluation Results
Method
Method
Links
Accuracy
Steps Taken
Agent-BRACE
Backbone=Qwen3-4B-Inst...
2026.05
88
30.8
PABU
Backbone=Qwen3-4B-Inst...
2026.05
83
26.4
ReAct (RL)
Backbone=Qwen3-4B-Inst...
2026.05
75.5
18.3
Direct-Action (RL)
Backbone=Qwen3-4B-Inst...
2026.05
74
29.8
Base Model
Backbone=Qwen3-4B-Inst...
2026.05
61.5
32.2
ReAct
Backbone=Qwen3-4B-Inst...
2026.05
61
12.8
Feedback
Search any
task
Search any
task