zai org GLM 5.1 needs about 461.6 GB of memory at Q4_K_M (a 459.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 1.1 GB to 801.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 5.6 GB.
zai org GLM 5.1 – 754 billion‑parameter mixture of experts model
zai org GLM 5.1 is a 754 billion‑parameter model built on the glm‑dsa architecture. It is a mixture of experts and was trained on 202 752 context tokens. The model is released under the MIT license.
Good at
- Strong coding and terminal task performance
- Long‑horizon agentic reasoning
- Large context handling
Limits
- Requires large VRAM for full precision quants
- No vision support
- Limited language metadata
Best for
- Complex coding projects
- Extended agentic workflows
- Terminal‑based automation
Who it’s for. Developers and researchers needing a large, expert‑based language model for extended reasoning tasks. The model is suitable for those who can allocate sufficient GPU memory and are comfortable with llama.cpp or similar frameworks.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.1 GB | 3.0 GB | Fits | Get file → | |
| IQ1_S | 158.2 GB (4 parts) | 160.1 GB | Too big | Get file → | |
| IQ1_M | 176.0 GB (5 parts) | 177.9 GB | Too big | Get file → | |
| IQ2_XXS | 203.9 GB (6 parts) | 205.8 GB | Too big | Get file → | |
| IQ2_XS | 226.6 GB (6 parts) | 228.5 GB | Too big | Get file → | |
| IQ2_S | 231.4 GB (6 parts) | 233.3 GB | Too big | Get file → | |
| IQ2_M | 254.9 GB (7 parts) | 256.8 GB | Too big | Get file → | |
| Q2_K | 265.8 GB (7 parts) | 267.7 GB | Too big | Get file → | |
| Q2_K_L | 266.8 GB (7 parts) | 268.7 GB | Too big | Get file → | |
| IQ3_XXS | 315.0 GB (9 parts) | 316.9 GB | Too big | Get file → | |
| Q3_K_S | 328.3 GB (9 parts) | 330.2 GB | Too big | Get file → | |
| IQ3_XS | 343.9 GB (9 parts) | 345.8 GB | Too big | Get file → | |
| Q3_K_M | 344.6 GB (9 parts) | 346.5 GB | Too big | Get file → | |
| Q3_K_L | 358.9 GB (10 parts) | 360.8 GB | Too big | Get file → | |
| Q3_K_XL | 359.7 GB (10 parts) | 361.6 GB | Too big | Get file → | |
| IQ3_M | 360.3 GB (10 parts) | 362.2 GB | Too big | Get file → | |
| IQ4_XS | 403.4 GB (11 parts) | 405.3 GB | Too big | Get file → | |
| IQ4_NL | 426.6 GB (11 parts) | 428.5 GB | Too big | Get file → | |
| Q4_0 | 427.6 GB (11 parts) | 429.5 GB | Too big | Get file → | |
| Q4_K_S | 441.4 GB (12 parts) | 443.3 GB | Too big | Get file → | |
| Q4_K_M | 459.7 GB (12 parts) | 461.6 GB | Too big | Get file → | |
| Q4_K_L | 460.4 GB (12 parts) | 462.3 GB | Too big | Get file → | |
| Q4_1 | 472.7 GB (13 parts) | 474.6 GB | Too big | Get file → | |
| Q5_K_S | 520.1 GB (14 parts) | 522.0 GB | Too big | Get file → | |
| Q5_K_M | 538.6 GB (14 parts) | 540.5 GB | Too big | Get file → | |
| Q6_K | 652.5 GB (17 parts) | 654.4 GB | Too big | Get file → | |
| Q8_0 | 801.3 GB (21 parts) | 803.3 GB | Too big | Get file → |
For 8 GB VRAM, use the IQ1_S quant (1.13 GB file, 3 GB memory estimate). For 16 GB VRAM, IQ2_XS (226.6 GB file, 228.5 GB memory estimate) or Q4_K_S (441.42 GB file, 443.3 GB memory estimate) are suitable. For 32 GB VRAM, Q4_K_L (460.38 GB file, 462.3 GB memory estimate) or Q5_K_M (538.61 GB file, 540.5 GB memory estimate) provide higher quality.
What GPU do you need for zai org GLM 5.1?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 3.0 GB in use).
Which zai org GLM 5.1 quant should you choose?
Q4_K_M is the usual starting point for zai org GLM 5.1: it balances file size and quality well. Here it is a 459.7 GB file.
The Q8_0 file is 801.3 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is the first one in the table at 1.1 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about zai org GLM 5.1
How much VRAM does zai org GLM 5.1 need?
About 461.6 GB at Q4_K_M with an 8K context window. That is the 459.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run zai org GLM 5.1 on an 8 GB GPU?
Yes, with the smaller files: the 1.1 GB file needs about 3.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of zai org GLM 5.1 GGUF?
It depends on the quant: file is 1.1 GB, IQ1_S is 158.2 GB, IQ1_M is 176.0 GB, IQ2_XXS is 203.9 GB, IQ2_XS is 226.6 GB, IQ2_S is 231.4 GB and 21 more.
What context length does zai org GLM 5.1 support?
The GGUF header lists a trained context length of 202,752 tokens (198K). The KV cache grows with the context you use: about 1.4 GB at 8K and 5.6 GB at 32K.
Where can I download zai org GLM 5.1 GGUF?
The files are on Hugging Face in the bartowski/zai-org_GLM-5.1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does zai org GLM 5.1 use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Yes, the IQ1_S quant uses 1.13 GB file size and estimates 3 GB VRAM usage at 8 k context.
Does this model support vision tasks?
No, the model has no vision capabilities.
What license does it use?
The model is released under the MIT license.