zai-org/GLM-4.7

GLM-4.7 is a 358 billion parameter language model developed by zai-org, featuring a 32768 token context length. It is specifically optimized for advanced agentic coding, terminal-based tasks, and complex reasoning, showing significant gains on benchmarks like SWE-bench and HLE. The model also excels in tool usage and improving UI quality for web pages and slides, making it suitable for sophisticated development and automation workflows.