GLM-5.2 — local AI model by Zhipu AI (Z.ai)
Written by Jakub Rusinowski · Last updated
Z.ai's June 2026 flagship and, by Artificial Analysis' ranking at the time, the strongest open-weight model available — fourth overall including closed models. A 744B-parameter Mixture-of-Experts with roughly 40B active per token and a genuinely usable 1M-token input context (output up to 131,072). Two selectable reasoning-effort levels. Released in stages between 13 and 17 June 2026: the Coding Plan first, then open weights on Hugging Face, then documentation. The MIT license is the notable part — full commercial use, modification and redistribution on a frontier-scale model.
Variants
The smallest GLM-5.2 variant needs about 450 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| GLM-5.2 744B → 744B (~40B active) | ~450 GB | 1,000,000 | ollama run glm-5-2 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run GLM-5.2 locally
Install Ollama, then pull the tag.
ollama run glm-5-2Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: GLM-5.2 744BRecommended GPU
The cheapest catalogued GPU that runs GLM-5.2 locally (min 450 GB VRAM) is the Apple M3 Ultra (512 GB).
Can I run GLM-5.2 on my GPU?
GLM-5.2 — frequently asked questions
How much VRAM does GLM-5.2 need?
GLM-5.2 needs about 450 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: GLM-5.2 744B (450 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run GLM-5.2 on an RTX 4090 (24 GB)?
GLM-5.2's smallest variant needs about 450 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for GLM-5.2?
Q4_K_M is the best balance of quality and VRAM for GLM-5.2 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run GLM-5.2 with Ollama?
GLM-5.2 has no local Ollama tag — the published tag is cloud-hosted, so running it sends your prompts to a hosted GPU rather than your own machine.