# GLM-5.3: hardware requirements
> GLM-5.3 (Z.ai) is a 744B (40B active) mixture-of-experts model with a 1.0M-token context window, released under the GLM-5.3 custom license. It needs about 466GB of memory at Q4_K_M with 8K context.
GLM-5.3 is Z.ai's flagship, released as open weights in August 2026, and it scores 44 on the Artificial Analysis Intelligence Index v4.3, level with Kimi K3 and among the highest of any open-weight model. It is a 744B mixture-of-experts with 40B active parameters, a 1M-token context, and a thinking mode, aimed at agentic coding and long-horizon tasks. At Q4_K_M it needs roughly 465GB with 8K context, so local use realistically means a 512GB Mac Studio or a multi-GPU server; the smallest community GGUFs start above 200GB. Unlike earlier GLM releases it uses a custom license with a revenue threshold, so check the terms before commercial use.

## Recommended hardware
- **Mac / unified memory:** M3 Ultra 512GB (512GB), Q3_K_M, ~13 tokens/sec

## Memory by quantization
| Quant | Weights | Total @ 8K ctx | Total @ 32K ctx |
| --- | --- | --- | --- |
| Q2_K | 279.0GB | 288.7GB | 290.9GB |
| Q3_K_M | 362.7GB | 374.9GB | 377.1GB |
| Q4_K_M | 451.0GB | 465.9GB | 468.1GB |
| Q5_K_M | 530.1GB | 547.3GB | 549.5GB |
| Q6_K | 613.8GB | 633.6GB | 635.8GB |
| Q8_0 | 790.5GB | 815.6GB | 817.8GB |
| BF16 | 1488.0GB | 1534.0GB | 1536.2GB |

## GPUs that run it fully in memory (fastest first)
| GPU | Memory | Quant | Speed | Price |
| --- | --- | --- | --- | --- |
| M5 Ultra 512GB | 512GB | Q3_K_M | ~19 tok/s | n/a |
| M3 Ultra 512GB | 512GB | Q3_K_M | ~13 tok/s | $9,499 MSRP |

## How to run it
- llama.cpp: `llama-server -hf unsloth/GLM-5.3-GGUF:Q4_K_M -c 8192 -ngl 99`
- Weights: https://huggingface.co/zai-org/GLM-5.3

## Strengths
- Among the strongest open-weight models for agentic coding and reasoning
- 1M-token context
- Toggleable thinking mode

## Weaknesses
- Needs roughly 465GB at Q4_K_M; even 1-bit to 2-bit builds exceed 200GB
- 40B active parameters make it slow on unified-memory machines
- Custom license with a revenue threshold, unlike the MIT-licensed GLM-5.3-Flash
- Text only
Full page: https://llamabuilds.ai/models/glm-5-3 · Hardware guide: https://llamabuilds.ai/models/glm-5-3/requirements

---
Source: https://llamabuilds.ai/models/glm-5-3 · Updated October 14, 2026 · Llama Builds. Speeds marked ~ are estimates from our bandwidth model (https://llamabuilds.ai/methodology); measured results are labeled as such.