Qwen4-Max Tested: 12 Points Below GPT-6.1 Sol at 2x the Price

WorldofAI · youtube · 2026-11-16

WorldofAI published a hands-on review of Alibaba's new flagship Qwen4-Max, which claims advanced reasoning, multimodal capabilities, a 1M-token context window, and agentic workflows.

Testing an early checkpoint on his own WOAIBench against GPT-6 Sol, Claude Opus 5.5, Gemini 4, and Kimi K3.1 Pro, the reviewer found Qwen4-Max scores 12 points below GPT-6.1 Sol on AA Index while costing 2x more. His verdict: not a bad model, but a disappointing flagship release relative to the official demos. He also notes future Qwen 4 models will be open source.

Original post →

More from Models

Models channel →