GLM 5.1

Zai's flagship foundation model designed for long-horizon autonomous tasks. GLM-5.1 achieves 58.4 on SWE-Bench Pro, surpassing GPT-5.4, with overall performance aligned with Claude Opus 4.6. Built for sustained autonomous operation on single tasks for up to 8 hours, completing the full loop from planning and execution to iterative optimization. With 203K context, 128K output, thinking capabilities, and tool support, it excels at complex system engineering requiring depth, persistence, and production-grade deliverables.