Qwen3.8 Max Ranks First Overall on Agentic Index
Key point
Qwen3.8 Max has taken the top spot overall on Artificial Analysis's Agentic Index.
Details
Qwen3.8 Max has recorded the #1 overall ranking on Artificial Analysis's Agentic Index.
Artificial Analysis compares the intelligence levels of major AI models by aggregating the following 9 evaluations in the Intelligence Index v4.1.1.
- GDPval-AA v2
- τ³-Banking
- Terminal-Bench v2.1
- SciCode
- Humanity’s Last Exam
- GPQA Diamond
- CritPt
- AA-Omniscience
- AA-LCR
In this update, τ³-Banking was changed to v1.0.1, and the graders for the HLE, AA-LCR, and AA-Omniscience evaluations were upgraded to GPT-5.6 Luna (medium).
The site also provides intelligence scores, output speed, and cost per Intelligence Index task for each model, supporting comparisons between open-weight and proprietary models as well as Pareto analysis of performance and cost. Additionally, it newly released the Endpoint Accuracy Index, which measures quality differences between endpoints serving the same model.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.