Over the course of the previous 24 hours, Phala Confidential AI successfully handled a total of 37.11 billion billed input and output tokens. This system operates as a TEE-backed AI infrastructure.
When observing the most frequently used models during this timeframe, z-ai/glm-5.2 ranked highest by processing 41.08% of the volume. Following closely was the deepseek/deepseek-v4-flash-0731 model with a 32.31% share, while moonshotai/kimi-k3 captured 8.70% of the usage.
For further details, you are welcome to visit https://phala.com/models/z-ai/glm-5.2
When observing the most frequently used models during this timeframe, z-ai/glm-5.2 ranked highest by processing 41.08% of the volume. Following closely was the deepseek/deepseek-v4-flash-0731 model with a 32.31% share, while moonshotai/kimi-k3 captured 8.70% of the usage.
For further details, you are welcome to visit https://phala.com/models/z-ai/glm-5.2