GLM-5.2 744B — VRAM, speed & local setup
Written by Jakub Rusinowski · Last updated
The full GLM-5.2 MoE. At 744B total parameters this needs roughly 400 GB even at Q4 — a multi-node or datacenter deployment, not a workstation, despite the modest ~40B active count. Reach it through the Z.ai API or a hosted provider unless you have that hardware. The 1M-token context and MIT license are what set it apart from other models at this scale.
GLM-5.2 744B needs about 450 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.
VRAM and speed by quantization
Quoted against NVIDIA RTX 4090 (24 GB). Weights plus overhead only: this model's architecture is unpublished, so no KV cache is added.
| Quant | Memory | VRAM | Speed (est.) | Fit |
|---|---|---|---|---|
| Q2_K 2.63 bpw | 245.4 GB | — | Too big | |
| Q3_K_M 3.41 bpw | 317.9 GB | — | Too big | |
| Q4_K_M 4.83 bpw | 450 GB | — | Too big | |
| Q5_K_M 5.67 bpw | 528.1 GB | — | Too big | |
| Q6_K 6.56 bpw | 610.9 GB | — | Too big | |
| Q8_0 8.50 bpw | 791.3 GB | — | Too big | |
| F16 16.00 bpw | 1488.8 GB | — | Too big |
Black marker = usable memory on the NVIDIA RTX 4090 (24 GB). Estimates from the memory-bandwidth roofline on the methodology page. GLM-5.2 744B VRAM calculator →
Get GLM-5.2 744B running
GLM-5.2 744B is not practical on a single consumer GPU: it needs about 450 GB at Q4_K_M, more than any single device in our catalogue holds with headroom. How multi-GPU inference works, or rent a multi-GPU server by the hour:
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
How to run GLM-5.2 744B
Install Ollama, then run:
ollama run glm-5-2Specifications
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
Preview — The model is released, but these specs are thin or rest on a single source. Individual fields may be wrong.
- Parameters
- 744 Billion (~40B active)
- Context window
- 1,000,000
- Architecture
- Mixture-of-Experts (two reasoning-effort levels)
- Provider
- Zhipu AI (Z.ai)
- Licence
- MIT
- Practical deployment
- Multi-GPU server
- Specified at
- Q4_K_M
- System RAM
- 768 GB
- Record updated
- 2026-08-15
Commercial use permitted. No usage restrictions beyond attribution.
Quality and use cases
Scores as published by the model’s authors or an independent evaluator — quality, not throughput, and not measured by us.
Can I run GLM-5.2 744B on my GPU?
GLM-5.2 744B — frequently asked questions
How much VRAM does GLM-5.2 744B need?
About 450 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does GLM-5.2 744B run on an RTX 4090 (24 GB)?
No. GLM-5.2 744B needs about 450 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run GLM-5.2 744B locally?
Install Ollama and run `ollama run glm-5-2`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.