GLM-5 Reasoning
GLM-5 Reasoning is an open-source AI model that achieves a top score of 85 on BenchLM and outperforms GPT-5.4 and Claude Opus 4.6 on the SWE-Bench Pro benchmark.
GLM-5 Reasoning is an open-source AI model that achieves a top score of 85 on BenchLM and outperforms GPT-5.4 and Claude Opus 4.6 on the SWE-Bench Pro benchmark.