Tech Mar 17, 2026 5 min GLM-OCR (0.9B) sets a new SOTA for document parsing, so I checked columns, vertical text, and math support Zhipu AI's GLM-OCR reaches 94.62% on OmniDocBench v1.5 despite using only 0.9B parameters. I dug into its layout parsing, vertical text handling, and math recognition. AI OCR VLM GLM