Zhipu AI

Zhipu AI is an AI company founded in 2019 and based in Beijing, on Zhongguancun East Road in Haidian District.¹ Its English-language site presents the company as Z.ai, under the name Z.AI Co., Ltd., and its Hugging Face account, zai-org, is also named Z.ai.¹² Zhipu says it developed the GLM pre-training framework at the end of 2020, trained the 10-billion-parameter GLM-10B in 2021, open-sourced the 130-billion-parameter bilingual GLM-130B in September 2022 and launched ChatGLM and the open ChatGLM-6B in March 2023, followed by GLM-4 in January 2024 and GLM-4.5, GLM-4.6 and GLM-4.7 during 2025.¹ It says its open-source model series has passed 40 million global downloads.¹ It sells API access through its Z.ai developer platform and runs the Z.ai chatbot, which the site says is powered by GLM-5.3-Flash.³⁴

Its models appear across MetaPrompt's launch posts. GLM 5 resolves 72.8% of SWE-bench Verified tasks in the shared mini-SWE-agent v2.0.0 harness, in a run dated February 17, 2026, level with GPT 5.2 in SWE-bench Verified, Pro and What Replaced Them: By the Numbers.⁵ GLM 5.3 Flash processed 12.67 trillion tokens on OpenRouter in the seven days to October 12, 2026, second only to DeepSeek V4.1 Flash, and GLM 5.3 was tenth with 3.96 trillion, as ranked in Most-Used AI Models on OpenRouter by Weekly Tokens.⁶ Z.ai lists GLM-5.3-Flash at $0.15 per 1M input tokens and $0.50 per 1M output tokens, tied for second in Cheapest LLM APIs per 1M Output Tokens, and GLM-5.3 at $1.40 and $4.40.³ GLM-5.3-Flash, under the MIT License, logged 6,511,401 Hugging Face downloads in the 30 days to October 13, 2026, second in the per-lab table of Chinese Open-Weight Models: Downloads, Derivatives and Token Share, while the full-size GLM-5.3 uses a GLM-5.3 License under which a Model-as-a-Service business above $10 billion in 12-month revenue must first pass Z.ai's security review.⁷