Expired70% confidenceFactExact time
GLM-5.1 had a 200,000-token context window, while GLM-5.2 multiplied that by 5x to reach 1 million tokens.
4
Sources
70%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
GLM-5.2 Deep Dive: Million-Token Context, MIT Open Source & Full-Stack Domestic Chip Training
bilibiliAI技术投降派6/14/2026
Related Claims
UnverifiedGLM-5.2 首次将长程任务能力建立在稳定可用的 100 万 token 上下文之上73% similarUnverifiedGLM-5.2 has 753 billion parameters, and its complete weight file exceeds 1.5 TB when stored in BF16 precision67% similarVerifiedGLM-5.2's model weights exceed 1.5 TB in download size, and even the 1-bit quantized compressed version requires approximately 200 GB of VRAM to run.64% similarUnverifiedThe context window of LLMs has expanded from the early 4K tokens to hundreds of thousands of tokens, enabling models to maintain state across complex multi-step tasks64% similarUnverifiedEarly mainstream models had context windows of only 4,000 to 8,000 tokens, while models from the past two generations have expanded the window to 100,000 tokens or higher.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/45327API
curl https://kongchang.com/api/v1/knowledge/claims/45327MCP
get_claim(id=45327)