Qwen3.7 Max Improves Reasoning and Coding Performance

·2026.05.22 19:45

Key point

Qwen3.7 Max's intelligence index rose compared to the previous model, showing major advances especially in scientific reasoning and coding.

Details

According to Artificial Analysis's analysis, Qwen3.7 Max recorded an Intelligence Index of 56.6 points, up about 5 points from the previous version, Qwen3.6 Max.

The performance improvements were mainly concentrated in scientific reasoning, agentic capability, and coding, with detailed benchmark metrics as follows:

CritPt: +9.7 p.p (3.7% → 13.4%)
HLE: +9.2 p.p (28.9% → 38.1%)
TerminalBench Hard: +6.9 p.p (43.9% → 50.8%)
GDPval-AA: +42 Elo (1504 → 1546)

It also showed the lowest hallucination rate among the models tested.

qwen artificial-analysis llm-benchmark ai-performance

Read originalShare on Threads

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.