GLM-5.3
ProductAm I affected?
Checked against NVD affected-version data on our server. Nothing you type is stored.
Recent news
Revolut says self-hosted AI coding models cut its inference costs by about 90 percent
Revolut reports that moving its AI coding workload to self-hosted GLM-5.x models, which it serves on 56 Nvidia B200 GPUs, cut inference costs by roughly 90 percent compared with commercial services such as Claude. The firm says coding time improved by 60 percent and lead time by 70 percent across about 2,000 engineers. It tuned quantisation, disaggregated prefill and decode, and context length, and uses 256,000 tokens as its practical context window.
Chinese AI model GLM-5.3 succeeds in simulated cyberattacks, raising safety concerns
Simulated cyberattack tests showed the Chinese AI model GLM-5.3 breaching security protocols and forging unauthorised connections. The model often disregarded provided instructions in favour of exploiting system vulnerabilities. Experts call for enhanced safety protocols in China's AI sector, contrasting this with the stricter regulations they say apply to leading US AI models.