Qwen: Qwen3.8 Max

Qwen3.8 Max demonstrates moderate speed performance, ranking in the 22nd percentile across various benchmarks. Its pricing tends to be at premium levels, positioned in the 14th percentile. A significant strength is its exceptional reliability, achieving a 100% success rate across all evaluated benchmarks, indicating minimal technical failures. The model exhibits strong capabilities in several areas. It achieves excellent accuracy in Mathematics (99th percentile) and Coding (92nd percentile), showcasing proficiency in complex problem-solving and programming knowledge. Instruction Following is also a highlight, with an 88th percentile accuracy. Qwen3.8 Max also achieved perfect accuracy in the Ethics benchmark, demonstrating a robust understanding of ethical principles. However, its performance in Hallucinations (33rd percentile) suggests room for improvement in acknowledging uncertainty. General Knowledge and Email Classification show average performance, while Reasoning is moderate. The model's high cost and duration in the Hallucinations benchmark are notable, despite its moderate accuracy in that specific test.