DeepSeek Expands AI Memory Tenfold Amid Zhipu AI's GLM-5 Launch
DeepSeek increases its AI model's context window to over 1 million tokens, enabling much more complex reasoning. Rival Zhipu AI simultaneously releases its upgraded GLM-5 model to keep pace in the competitive Chinese AI market.
Chinese artificial intelligence start-up DeepSeek expands its flagship AI model by increasing its context window from 128,000 tokens to over 1 million. This nearly tenfold upgrade allows the AI system to remember and process significantly more information within a single conversation or task, enabling more complex reasoning and better handling of data and code.
The major update arrives on the exact same day that fellow Chinese AI heavyweight Zhipu AI unveils its next flagship model, GLM-5. This simultaneous release intensifies the ongoing competition among Chinese technology firms to develop the most capable artificial intelligence systems.
Zhipu AI's new GLM-5 model features a twofold increase in parameters and delivers improved coding and agentic capabilities. Notably, the model adopts DeepSeek Sparse Attention, an efficiency technique invented by DeepSeek that balances high AI performance with computational cost.