WebChallenger: A Reliable and Efficient Generalist Web Agent
Open paper
Benchmark
Model
Rank
Results
computer-use-agents-on-visualwebarena
WebChallenger (GLM-4-32B)
#1
Success Rate: 48.7
computer-use-agents-on-webarena
WebChallenger (GLM-4-32B)
#2
Success Rate: 56.3
Rank counts only results with a code link.