Z.ai's GLM-5.2 challenges GPT-5.5 on coding benchmarks

Chinese AI startup Z.ai has released GLM-5.2, a 753-billion parameter open-weights large language model designed for autonomous coding tasks. This model boasts a 1-million-token context window and offers enterprise subscription tiers starting at a low monthly cost. Its weights are available under an MIT open-source license, allowing for customization and local deployment, which appeals to businesses concerned about regulatory uncertainties and vendor lock-in with proprietary models. GLM-5.2 utilizes an 'IndexShare' optimization to reduce computational costs for long-context processing and features selectable 'Thinking Modes' for balancing performance and latency.