Zhipu Launches GLM-5.3 Flash to Compete with DeepSeek V4 Flash

Zhipu has introduced its latest model, the GLM-5.3 Flash, designed to compete directly with the DeepSeek V4 Flash. This new lightweight flagship model utilizes over 100,000 domestic chip clusters to deliver inference services. With a parameter scale of approximately 300 billion, the GLM-5.3 Flash features an MoE architecture that activates around 18 billion parameters, enabling native multimodal capabilities for processing both image and video inputs.

The GLM-5.3 Flash has received a comprehensive evaluation score that surpasses its predecessor, the GLM-5.2, and is comparable to the Claude Opus 4.8. It also outperforms the DeepSeek V4 Pro official version in relevant score metrics. The pricing structure for the model is set at 0.8 yuan per million tokens for input, 2.8 yuan for output, and 0.23 yuan for cache hits, which is approximately one-tenth of the cost of GLM-5.3, with a promotional half-price discount available for the first two weeks after launch.

Zhipu emphasizes that the GLM-5.3 Flash is fully compatible with domestic computing power, with online traffic managed by domestic chips. The company claims that the hardware efficiency and unit token cost of this model are on par with mainstream NVIDIA GPUs. Prior to its official release, the model was tested anonymously on overseas platforms under the code name Ox Alpha.

© 2026 KLEA News. All Rights Reserved. This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.

Source: KLEA News