--- title: "GLM 5.2: an open-weight frontier under an MIT licence, and why it matters for AI sovereignty" description: "Z.ai releases GLM 5.2 with open weights (MIT) and a 1M-token context. According to Artificial Analysis it's the best open-weight model and fourth overall. What it means for organisations that want frontier AI they can control and self-host." url: "https://www.noze.it/en/insights/glm-5-2-open-weight-ai-sovereignty/" lang: "en" alternate: "https://www.noze.it/insights/glm-5-2-open-weight-sovranita-ai/" published: "2026-06-20" author: "noze S.r.l." keywords: "AI, Open Source, Governance, Open Weight, LLM, GLM, AI Sovereignty, On-Premise, Coding" --- # GLM 5.2: an open-weight frontier under an MIT licence, and why it matters for AI sovereignty Z.ai releases GLM 5.2 with open weights (MIT) and a 1M-token context. According to Artificial Analysis it's the best open-weight model and fourth overall. What it means for organisations that want frontier AI they can control and self-host. 20 June 2026 ![Bar chart of the Artificial Analysis Intelligence Index: GLM-5.2 at 51, fourth overall behind Claude Fable 5 (60), Claude Opus 4.8 (56) and GPT-5.5 (55)](https://www.noze.it/images/insights/glm-5-2-intelligence-index.png) The Artificial Analysis Intelligence Index: GLM-5.2 scores 51, top of the open-weight field and fourth overall, eleven points above GLM-5.1. Image: Artificial Analysis (artificialanalysis.ai). ## What GLM 5.2 is Between 13 and 17 June 2026, **Z.ai** (formerly Zhipu AI) released **GLM 5.2** in stages: first on the **GLM Coding Plan** (13 June), then by publishing the **open weights** on Hugging Face (16 June), and finally with documentation and benchmarks (17 June). It is the direct successor to GLM-5.1 in the GLM line (4.5 → 4.6 → 5 → 5.1 → 5.2). The headline characteristics, derivable from the model card and the `config.json` shipped with the weights: - **Mixture-of-Experts architecture** (not _dense_): **around 753 billion total parameters**, of which **\~40 billion active** per token. The weight files are roughly **1.5 TB** in BF16 (about 750 GB in the FP8 variant). - **1-million-token context** (it was 200K in GLM-5.1), with output up to \~128-131K tokens. - A **reasoning model** with two “thinking effort” levels; **text-only** (no multimodal input). - **MIT licence**, verified on the repository’s `LICENSE` file, with weights downloadable from Hugging Face (`zai-org/GLM-5.2`, plus an FP8 variant). Permissive: commercial use, modification, fine-tuning, redistribution and self-hosting, with no fees and no use restrictions. - **API and pricing** on z.ai: roughly **$1.40/million** input tokens and **$4.40/million** output (the same as GLM-5.1), with the GLM Coding Plan starting at a few tens of dollars a month. The stated positioning is clear: **high-performance coding and long-horizon agentic tasks**, with an Anthropic-compatible endpoint, so it works day one in environments like Claude Code, Cline and OpenCode. ## How strong it is (and who says so) This calls for a distinction we always make: separate the numbers **reported by the vendor** from **independent** measurements. The strongest independent data point comes from **Artificial Analysis**: in their Intelligence Index, GLM 5.2 scores **51**, making it **the best open-weight model** on the board, seven points clear of the next open model (MiniMax-M3 and DeepSeek V4 Pro at 44, Kimi K2.6 at 43) and an **+11-point** jump over GLM-5.1. Overall it lands **fourth**, behind only three _closed_ models: Claude Fable 5 (60), Claude Opus 4.8 (56) and GPT-5.5 (55). Also from independent evaluations, it ranks second on Code Arena for frontend development tasks, and analyst Simon Willison calls it “probably the most capable open-weight text-only LLM”. Z.ai then publishes its own benchmarks (for example SWE-bench Pro 62.1, GPQA-Diamond 91.2, Terminal-Bench 81.0), but these read as **self-reported figures**: Artificial Analysis’s independent _re-runs_ return lower values (GPQA-Diamond around 89%, Terminal-Bench around 78%), owing to different scaffolding and test configurations. Two honest caveats from those same independent sources: the model is **“token-hungry”** (it burns far more reasoning tokens per task than its predecessor), and it shows some **regression on creative tasks** versus GLM-5.1. The picture, in short, is a very strong coding-and-agentic model, not blanket supremacy on every metric. What made the headlines stands: coding performance in the same bracket as Western frontier models at a **fraction of the cost** (the press cites roughly one-sixth of GPT-5.5 on some long-horizon benchmarks). ## The real point for us: a frontier you can own The detail that truly matters isn’t a benchmark, it’s the licence. **GLM 5.2 is an open-weight frontier model under MIT**: you download the weights, run them wherever you want, and nobody can switch them off. That is exactly the argument we made a few days ago, recounting how [a government was able to disable a commercial model](https://www.noze.it/en/insights/fable-5-shutdown-us-government-ai-sovereignty/) overnight. The model at the top of that independent ranking, Claude Fable 5, is the one that was disabled. Weights held locally are not exposed to that decision. A bit of technical honesty, though: “open-weight” doesn’t mean “runs on a laptop”. It needs datacentre-class resources: about **8 H200 GPUs** for the FP8 version (\~750 GB), twice that in BF16. Community quantisations (GGUF for llama.cpp, Ollama, LM Studio) lower the bar at the cost of quality, but full self-hosting is realistic for those with a GPU cluster or rented cloud capacity. For many teams the managed API stays the pragmatic choice; self-hosting earns its keep when you have **compliance, data-residency or sustained-throughput** requirements. It’s the same reasoning behind our [on-premise AI workstations](https://www.noze.it/en/insights/gb10-ai-workstation-on-premise/) and our use of runtimes like [vLLM](https://www.noze.it/en/insights/vllm-open-source/) and [Ollama](https://www.noze.it/en/insights/ollama-open-source/). ## Sovereignty and compliance: local weights ≠ a China-hosted API There’s a distinction a European business or public body must hold firm, because it’s legal in nature, not about model quality. - **Running the weights** locally or in an EU private cloud means prompts, context and data **stay on your own infrastructure**: no sub-processor to vet, no cross-border transfer to document. _Data residency_ becomes a property of the architecture. The MIT licence, moreover, imposes no regional or use restrictions. - **Using the Z.ai-hosted API** (in China) is a different risk profile: data transits a Chinese jurisdiction subject to the _National Intelligence Law_ (2017) and the **PIPL** transfer regime. For an EU controller this is an extra-EU transfer to assess under the **GDPR** (Articles 44 ff.), with a DPIA and an adequate legal basis. It’s also worth recalling that **Zhipu AI has been on the US Department of Commerce Entity List since January 2025**: it doesn’t prevent running MIT weights locally, but it’s a due-diligence factor for anyone routing data to the API. This is exactly the kind of assessment we run in [AI governance](https://www.noze.it/en/insights/open-intelligence-secure-governance-agentic/) projects. The operational takeaway is that, with open weights, **the choice goes back into the organisation’s hands**. ## What to do - **Evaluate it where it excels.** For coding and agentic workflows, GLM 5.2 is now a serious candidate, especially on price/performance. Test it on _your_ tasks, not on benchmarks. - **Decide weights vs API by compliance.** Sensitive data, public sector, regulated industries → self-host or an EU-residency endpoint. Non-critical workloads → the managed API may suffice, with the due diligence above. - **Don’t tie yourself to a single model.** A _gateway_ with multi-vendor fallback makes GLM 5.2 a routing choice, not a hard dependency. - **Ride the open-weight wave.** GLM sits alongside [DeepSeek V4](https://www.noze.it/en/insights/deepseek-v4-preview/), [Qwen](https://www.noze.it/en/insights/qwen-3-6/) and [Gemma](https://www.noze.it/en/insights/gemma-new-release/): a plurality of downloadable, competitive models is itself a hedge against any single provider’s _kill-switch_. It’s the difference between being subject to a roadmap and governing one: the point of our work on [artificial intelligence](https://www.noze.it/en/artificial-intelligence/), of the open-source [Admina](https://www.noze.it/en/open-source/admina/) framework and of products like [IntelliPA](https://www.noze.it/en/products/intellipa/). ## Where we are As of this writing, **20 June**, GLM 5.2 is the open-weight reference point of the moment: the best independent score in its category and the first open model to close in on the _closed_ frontier. Some independent evaluations on benchmarks Z.ai hasn’t yet published (SWE-bench Verified, LiveCodeBench, Aider) still need to settle; as always, the numbers should be confirmed in the field. But the underlying point is already clear: frontier intelligence, now, is something you can also **own**. **Link**: [Z.ai: GLM 5.2 blog](https://z.ai/blog/glm-5.2) · [Hugging Face: zai-org/GLM-5.2](https://huggingface.co/zai-org/GLM-5.2) · [Artificial Analysis](https://artificialanalysis.ai/articles/glm-5-2-is-the-new-leading-open-weights-model-on-the-artificial-analysis-intelligence-index) · [Simon Willison](https://simonwillison.net/2026/Jun/17/glm-52/)