MiniMax AI's M3 model has achieved the seventh position on the Long-Horizon Terminal-Bench, a benchmark designed to evaluate model performance on extended tasks. Additionally, an open-weight version of the model secured the top spot among open-weight models tested on the same benchmark. The Long-Horizon Terminal-Bench, established around May, previously indicated that even the top-performing models could only complete seven out of 46 tasks, with an average completion rate of two tasks across all tested models. AI
IMPACT Demonstrates advancements in long-horizon task performance for AI models, potentially influencing future model development and evaluation.
RANK_REASON The cluster reports on benchmark performance of AI models, which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
