Z.AI: GLM 4.7 Flash

Z.AI's GLM-4.7-Flash, a 30B-class model released on January 19, 2026, is positioned as a balanced option for performance and efficiency, particularly optimized for agentic coding. While it demonstrates exceptional reliability with a 97% success rate, indicating consistent operational stability, its speed performance tends to be slower, ranking in the 2nd percentile across benchmarks. Pricing is moderate, falling within the 39th percentile. In terms of specific benchmark results, GLM-4.7-Flash achieved perfect 100.0% accuracy in Email Classification, notably being the most accurate model at its price point and among models of comparable speed. This highlights a significant strength in classification tasks. Its performance on the Ethics benchmark was solid at 98.0% accuracy. However, a notable weakness is its 67.3% accuracy on the Hallucinations (Baseline) test, placing it in the 17th percentile, suggesting room for improvement in acknowledging uncertainty for fictional concepts. The model's duration metrics across all benchmarks indicate longer response times, reinforcing its lower speed ranking.