AIHacker Newsabout 2 hours ago
AutoBot Achieves Top Score on AI Benchmark
12 min read
AutoBot, an open-source agentic harness, achieved 50.70% accuracy on the official AssistantBench hidden-test leaderboard, surpassing published baselines for OpenAI and Anthropic models on complex, multi-application workflows.
Level
Hype check
Tap to vote and see what everyone thinks.
