Z.ai publishes GLM-5.2 open weights under MIT on Hugging Face

Category: AI
Status: confirmed
Sources: Hugging Face zai-org/GLM-5.2, discussion
Summary: Z.ai (Zhipu) published the GLM-5.2 open-weight checkpoint on Hugging Face under the MIT license, fulfilling the open-weight release promised at the 2026-06-13 announcement. The repository lists a 753B-parameter mixture-of-experts model in BF16 and F32 tensors with a 1M-token context window and an "IndexShare" attention design that reuses one indexer across every four sparse attention layers. Community quantized variants for llama.cpp, Ollama, and LM Studio appeared within hours. Vendor and secondary coding-benchmark numbers remain unreproduced and should be treated as developing.
Why it matters: An MIT-licensed model that topped the open-weights intelligence index is now runnable locally, removing the China-hosted-API data-handling concern and raising migration pressure on paid coding and agent APIs.
Follow-up: Track independent coding-benchmark reproduction and adoption in agent harnesses.