Alibaba's Qwen 3.8-Max, marketed as outperforming GPT-5.6, Sol Max, and Fable 5 on agentic computer use, ranked mid-pack or last in independent VulcanBench tests, a discrepancy driven by time budgets:…
IntelligenceLab requests Hugging Face to add Long-Horizon-Terminal-Bench (LHTB) to the Benchmark allow-list. LHTB is a 46-task benchmark that tests LLM agents on long-horizon terminal tasks using hidd…
A new benchmark, Long-Horizon-Terminal-Bench, tests AI agents on 46 long-horizon tasks averaging 231 episodes and 85.3 minutes per run, with top models achieving only a 15.2% pass rate at a 0.95 parti…
Researchers introduced Long-Horizon-Terminal-Bench, a benchmark of 46 long-horizon tasks across nine categories including experiment reproduction and software engineering, designed to evaluate AI agen…