UTC

GLM-5.2

Zhipu AI (Z.ai) · open weights · served by 134 providers · tool calling, reasoning

zhipuai/glm-5.2

Weights
Open
Released
2026-06-13
Knowledge cutoff
Not stated
Context window
1M
Deprecation dates

When each platform deprecates GLM-5.2, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.

  • Databricksdeprecates 2027-02-04databricks/databricks-glm-5-2successor: databricks/databricks-glm-5-3
  • aianddeprecated 2026-10-27aiand/zai-org/glm-5.2successor: aiand/zai-org/glm-5.3
  • Mistral APIdeprecated 2026-10-31mistral/zai-glm-5-2successor: mistral/zai-glm-5-3
  • scalewaydeprecated 2026-11-16scaleway/glm-5.2successor: zai/glm-5.3
Every upcoming deprecation →
Hugging Face repository

zai-org/GLM-5.2 at commit cf457fa734, last changed 2026-09-01

Licence
mit
Access
Open download
Library
transformers
Formats
.safetensors
  • No model signature file is published with the weights.

What the Hub's scanners report

39 weight files, 39 with a scan result. No scanner reports an issue with them.

The model at this commit, its 39 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • 302.AIglm-5.2
  • Abacuszai-org/GLM-5.2
  • above.devglm-5.2
  • above.devglm-5.2-fast
  • Adaptionglm-5.2
  • AIHubMixcoding-glm-5.2-free
  • AIHubMixglm-5.2
  • AI/ML APIzhipu/glm-5.2
  • Alibabaglm-5.2
  • Alibaba (China)glm-5.2
  • Alibaba Token Planglm-5.2
  • Alibaba Token Plan (China)glm-5.2
  • Ambientambient/large
  • Ambientz-ai/glm-5.2
  • Ambientzai-org/GLM-5.2-FP8deprecated
  • Arceezai-org/glm-5.2beta
  • Basetenzai-org/GLM-5.2
  • Basetenzai-org/GLM-5.2-Fast
  • Berget.AIzai-org/GLM-5.2deprecated
  • Boundlessglm-5.2deprecated
  • BytePlus ModelArkglm-5-2-260617
  • Chuteszai-org/GLM-5.2-TEE
  • ClinePasscline-pass/glm-5.2deprecated
  • Cloudflare Workers AI@cf/zai-org/glm-5.2deprecated
110 more providers

CommandCode (zai-org/GLM-5.2), CommandCode (zai-org/GLM-5.2-Fast), Cortecs (glm-5.2), CrossModel (z-ai/glm-5.2), Crusoe (zai/GLM-5.2), Databricks (databricks-glm-5-2), Deep Infra (zai-org/GLM-5.2), DigitalOcean (glm-5.2), DInference (glm-5.2), Doubleword (zai-org/GLM-5.2-FP8), Eden AI (zai/glm-5.2), EmpirioLabs AI (glm-5-2), engy (glm-5.2), evroc (zai-org/GLM-5.2), Friendli (zai-org/GLM-5.2), GMI Cloud (zai-org/GLM-5.2-FP8), Vertex (zai-org/glm-5.2-maas), GreenPT (glm-5.2), GreenPT (glm-5.2-caveman), GreenPT (glm-5.2-caveman-lite), GreenPT (glm-5.2-caveman-ultra), GreenPT (glm-5.2-honey), GreenPT (glm-5.2-honey-lite), GreenPT (glm-5.2-honey-ultra), GreenPT (glm-5.2-ponytail), GreenPT (glm-5.2-ponytail-lite), GreenPT (glm-5.2-ponytail-ultra), HPC-AI (zai-org/glm-5.2), Huawei Cloud MAAS (glm-5.2), Hugging Face (zai-org/GLM-5.2), Impossibl (fireworks/glm-5.2), Impossibl (zai/glm-5.2), IteraCompute (z-ai/glm-5.2), Jalapeno Cloud (GLM-5.2), Keln (zai-org/glm-5.2), Kenari (glm-5-2), Kilo Gateway (z-ai/glm-5.2), Kilo Gateway (z-ai/glm-5.2-fast), DevPass (LLM Gateway) (glm-5.2), DevPass (LLM Gateway) (glm-5.2-fast), LLM Gateway (alibaba/glm-5.2), LLM Gateway (baidu/glm-5.2), LLM Gateway (bytedance/glm-5.2), LLM Gateway (canopywave/glm-5.2), LLM Gateway (flexai/glm-5.2), LLM Gateway (novita/glm-5.2), LLM Gateway (runware/glm-5.2), LLM Gateway (scx-ai-gp/glm-5.2), LLM Gateway (scx-ai-gp/glm-5.2-fast), LLM Gateway (tencent/glm-5.2), LLM Gateway (zai/glm-5.2), Merge Gateway (zai/glm-5.2), Mistral (glm-5-2), Mistral (zai-glm-5-2), Model Oracle AI (glm-5.2), Morph (morph-glm52-744b), NanoGPT (TEE/glm-5.2), NanoGPT (TEE/glm-5.2:thinking), NanoGPT (z-ai/glm-5.2), NanoGPT (z-ai/glm-5.2:thinking), Nebius Token Factory (zai-org/GLM-5.2), Neon (glm-5-2), Novita AI (zai-org/glm-5.2), Ofox (z-ai/glm-5.2), Ollama Cloud (glm-5.2), OpenCode (glm-5.2), OpenCode Go (glm-5.2), OpenRouter (z-ai/glm-5.2), Opper (glm-5.2), OrcaRouter (z-ai/glm-5.2), Phala (z-ai/glm-5.2), Pioneer (zai-org/GLM-5.2), Prime Inference (z-ai/glm-5.2), Regolo AI (glm5.2), Requesty (glm-5.2), Requesty (glm-5.2@eu), routing.run (glm-5.2), routing.run (glm-5.2-nitro), Scaleway (glm-5.2), SCNet Token Plan (GLM-5.2), SCX.ai (GLM-5.2), SCX.ai (GLM-5.2-Fast), SenseNova (China) (glm-5.2), SiliconFlow (zai-org/GLM-5.2), SiliconFlow (China) (zai-org/GLM-5.2), Standard Compute (z-ai/glm-5.2), Tencent Token Plan (glm-5-2), Tencent Token Plan (glm-5.2), TensorX (z-ai/glm-5.2), Together AI (zai-org/GLM-5.2), TokenGo (z-ai/glm-5.2), Tokenify (z-ai/glm-5.2), TrustedTokens (zai-org/GLM-5.2), UnoRouter (glm-5.2), UnoRouter (glm-5.2:free), Venice AI (zai-org-glm-5-2), Vercel AI Gateway (zai/glm-5.2), Vercel AI Gateway (zai/glm-5.2-fast), Vivgrid (glm-5.2), Volcengine Ark (glm-5-2-260617), Vultr (glm-5.2), Wafer (GLM-5.2), CoreWeave (zai-org/GLM-5.2), Watt Cloud (GLM-5.2), Z.AI (glm-5.2), Z.AI Coding Plan (glm-5.2), Z.AI Coding Plan (glm-5.2-highspeed), Zenifra (zenifra/glm-5.2), ZenMux (z-ai/glm-5.2), Zhipu AI (glm-5.2)

Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.