Qwen3.8 27B Scores 52 on Artificial Analysis Intelligence Index
Qwen3.8 27B, an open-weights model, achieves an Artificial Analysis Intelligence Index of 52, placing it among top performers. The index, based on nine evaluations including GDPval, Terminal-Bench, and GPQA Diamond, measures reasoning, knowledge, and agentic tool use. The model's performance is highlighted in a detailed breakdown of intelligence, cost, token usage, and context window, with comparisons against other models on cost and efficiency.
While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.