GLM-5.3 family: which size to run
2 sizes, 320B–744B parameters · Z.ai
Sizes
| Model | Architecture | |||||||
|---|---|---|---|---|---|---|---|---|
| GLM-5.3 Flash | 320B | MoE · 18B active | 1,048,576 | 71.6 | 248 | 242 | No GGUF Q4_K_M file (plain or UD-) in the repository collected | 204 |
| GLM-5.3 | 744B | MoE · 40B active | 1,048,576 | 76.1 | 566 | 560 | No GGUF Q4_K_M file (plain or UD-) in the repository collected | 418.3 |
Each size is its own model, with its own scores and memory needs. Q4 estimates are planning figures, not measurements; file sizes are published. The Mac or PC column is for a 32 GB or larger machine (8 GB reserve); machines under 32 GB reserve 25% of their memory (2 to 8 GB), so they need less. The publisher lists GLM-5.3 744B-A40B, GLM-5.3-Flash 320B-A18B.
GLM-5.3 Flash starts from a newly trained base model, so it is not GLM-5.3 made smaller: the two sizes differ in base model as well as in parameters.
How to read this table
Q4 means compressed to about 4 bits per weight. GGUF (llama.cpp, Ollama, LM Studio) and MLX (Apple silicon) are file formats; the sizes shown are the published 4-bit files. A dash means no value is published or collected for that size; the reason is in the cell's accessible text.
Which size fits my machine
The largest size whose Q4 planning estimate fits what each machine offers. A planning estimate, not a measurement and not a speed. Smaller machines are left out; the largest one that fits no size is kept to show the limit.
| Machine | Largest GLM-5.3 size with enough memory | ||
|---|---|---|---|
| Mac Studio M5 Max · 128 GB unified | 104 | No size of this family has a Q4 estimate within what this machine offers | No size fits, so there is no estimate to compare |
| Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unified | 392 | GLM-5.3 Flash | 248 |
| RTX 5090 · 32 GB VRAM | 32 | No size of this family has a Q4 estimate within what this machine offers | No size fits, so there is no estimate to compare |
| DGX H100 · 8 × 80 GB | 640 | GLM-5.3 | 574 |
A dash means no size of this family fits that machine. The requirement is the size's Q4 estimate with that machine's own reserve: a graphics card reserves 2 GB, as in the sizes table; a Mac or PC reserves 25% of its memory (2 to 8 GB), so below 32 GB its figure is lower than the sizes table's. A server profile reserves 2 GB per graphics card, so its figure is higher.
Smallest machine for each size
The profile with the least memory that still has enough for the size's Q4 estimate, from the hardware catalog.
| Size | Smallest Mac or PC-memory profile | Smallest graphics-card profile | Server-class profile |
|---|---|---|---|
| GLM-5.3 Flash | Mac Studio 2026 · M5 Ultra · 512 GB | No graphics-card profile in the catalog has enough memory for this size | DGX H100 · 8 × 80 GB · 640 GB |
| GLM-5.3 | No Mac, laptop or PC profile in the catalog has enough memory for this size | No graphics-card profile in the catalog has enough memory for this size | DGX H100 · 8 × 80 GB · 640 GB |
A dash means no profile of that kind in the catalog has enough memory for the size.
Memory by file precision
The same planning formula at three precisions (0.625, 1.125 and 2 bytes per parameter, plus working room and the system reserve). Estimates, not measurements.
| Size | ||||||
|---|---|---|---|---|---|---|
| GLM-5.3 Flash | 248 | 242 | 440 | 434 | 776 | 770 |
| GLM-5.3 | 566 | 560 | 1,013 | 1,007 | 1,794 | 1,788 |
Family tree
Family, sizes and, for each size, the two 4-bit files the tables use (GGUF Q4_K_M and MLX 4-bit) when one was collected. Other precisions are not drawn here.
The same tree as a list:
- GLM-5.3 · Z.ai
- Flash 320B A18B · 18B active
- MLX 4-bit · 204 GB · mlx-community/GLM-5.3-Flash-4bit, read Oct 23, 2026
- 744B A40B · 40B active
- MLX 4-bit · 418.3 GB · mlx-community/GLM-5.3-4bit, read Oct 23, 2026
- Flash 320B A18B · 18B active
Learn more
See GLM-5.3 Flash
The smallest size of the family: its scores, memory estimate and the machines it runs on.
Sources
Where the family and its sizes are named, and where each number comes from.
- zai-org/GLM-5 README · “Use GLM-5.3 & GLM-5.3-Flash API services on Z.ai API Platform.” · read Oct 17, 2026
- zai-org/GLM-5 README, model download table (rows GLM-5.3 and GLM-5.3-Flash) · “| GLM-5.3 | … | 744B-A40B | FP8 | … | GLM-5.3-Flash | … | 320B-A18B | FP8 |” · read Oct 17, 2026
- zai-org/GLM-5.3-Flash model card · “GLM-5.3-Flash starts from a newly trained base model, with its architecture and training recipe redesigned around capability and efficiency.” · read Oct 17, 2026
- zai-org/GLM-5.3 model card · “GLM-5.3 uses the same base model as GLM-5.2 — every gain comes from post-training.” · read Oct 17, 2026
- GLM-5.3 Flash model card
- GLM-5.3 model card
- LiveBench 2026-06-25 · the only quality source on this page; a dash means the release does not include the model · snapshot collected Sep 26, 2026
- Published file sizes: Hugging Face repositories, read Oct 23, 2026
- Methodology · how the Q4 planning estimate is computed · Learn