GLM-5.1

GLM-5.1 is Zhipu AI's next-generation flagship model for agentic engineering, succeeding GLM-5 with significantly stronger coding and long-horizon task capabilities. A 754B parameter Mixture-of-Experts with 40B active parameters per token, it achieves state-of-the-art performance on SWE-Bench Pro (58.4), outperforming GPT-5.4, Claude Opus 4.6, and Gemini 3.1 Pro. Designed for sustained autonomous work, GLM-5.1 can operate on a single task for up to 8 hours — planning, executing, testing, and iterating across hundreds of rounds and thousands of tool calls. Its MoE architecture keeps VRAM requirements manageable despite the massive parameter count, making it accessible on high-end consumer hardware at Q4_K_M quantization.