Long-Horizon Terminal-Bench: LLMs Still Struggle
iamfakhrealam · x · 2026-07-13
Long-Horizon Terminal-Bench tested 20 frontier models across 46 terminal tasks, with a 90-minute limit per task. Results show that even the best-performing model solved only 13 tasks, 29 tasks were unsolved by any model, and about 55% of runs scored below expectations. This indicates that LLMs still face significant challenges when handling long-horizon, real-world tasks.
Related event: Tencent Hunyuan Releases Long-Horizon Terminal Bench Exposing Model Limits(6 posts)→
More from Models
- MissionReport benchmark: only 3 of 540 AI-generated government reports pass all criteria — Exp_Mark · 2026-10-27
- Local beats cloud: Qwen 3.8 FN handles private student data as well as frontier models — dr_lm · 2026-10-27
- Aidan Gomez: model's CoT said no tool calls right before making them — aidangomez · 2026-10-27
- Nat Lambert: Cohere's model is narrow RL training on internal infra ill-suited for open weights — natolambert · 2026-10-27
- OpenAI teases DOTS is coming to Europe 'soon' — mark_k · 2026-10-27
- Clifford Pickover challenges AI to decipher a message he encoded in Norse runes — pickover · 2026-10-27