Glm-4.5-Air needs about 75.1 GB of memory at Q4_K_M (a 73.0 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 38.3 GB (TQ1_0) to 221.0 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 6.3 GB.
GLM‑4.5‑Air: 106 billion‑parameter hybrid reasoning model
GLM‑4.5‑Air is a compact foundation model with 106 billion total parameters and 12 billion active parameters. It supports hybrid reasoning with thinking and non‑thinking modes. The model is released under the MIT license and can be used commercially.
Good at
- Hybrid reasoning and tool usage
- Compact parameter count for efficient deployment
- Supports 131 072‑token context
Limits
- Large VRAM requirement for higher‑quality quants
- No vision or multimodal support
- Requires 8 k context for memory estimates
Best for
- Intelligent agent back‑ends
- Complex reasoning workflows
- Tool‑integrated chat applications
Who it’s for. Developers building intelligent agents who need reasoning and coding capabilities in a compact model. The model is suitable for commercial use under the MIT license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 38.3 GB | 40.3 GB | Too big | Get file → | |
| IQ1_S | 38.5 GB | 40.6 GB | Too big | Get file → | |
| IQ1_M | 40.1 GB | 42.1 GB | Too big | Get file → | |
| IQ2_XXS | 42.7 GB | 44.8 GB | Too big | Get file → | |
| IQ2_M | 44.4 GB | 46.5 GB | Too big | Get file → | |
| Q2_K | 45.3 GB | 47.4 GB | Too big | Get file → | |
| Q2_K_L | 45.5 GB | 47.5 GB | Too big | Get file → | |
| Q2_K_XL | 47.4 GB | 49.5 GB | Too big | Get file → | |
| IQ3_XXS | 51.4 GB (2 parts) | 53.5 GB | Too big | Get file → | |
| Q3_K_S | 52.7 GB (2 parts) | 54.7 GB | Too big | Get file → | |
| Q3_K_XL | 54.8 GB (2 parts) | 56.9 GB | Too big | Get file → | |
| Q3_K_M | 57.2 GB (2 parts) | 59.3 GB | Too big | Get file → | |
| IQ4_XS | 60.5 GB (2 parts) | 62.6 GB | Too big | Get file → | |
| IQ4_NL | 62.7 GB (2 parts) | 64.8 GB | Too big | Get file → | |
| Q4_0 | 62.8 GB (2 parts) | 64.9 GB | Too big | Get file → | |
| Q4_K_S | 67.0 GB (2 parts) | 69.1 GB | Too big | Get file → | |
| Q4_K_XL | 67.7 GB (2 parts) | 69.8 GB | Too big | Get file → | |
| Q4_1 | 69.4 GB (2 parts) | 71.5 GB | Too big | Get file → | |
| Q4_K_M | 73.0 GB (2 parts) | 75.1 GB | Too big | Get file → | |
| Q5_K_S | 78.4 GB (2 parts) | 80.5 GB | Too big | Get file → | |
| Q5_K_XL | 82.8 GB (2 parts) | 84.9 GB | Too big | Get file → | |
| Q5_K_M | 83.5 GB (2 parts) | 85.6 GB | Too big | Get file → | |
| Q6_K | 99.0 GB (2 parts) | 101.1 GB | Too big | Get file → | |
| Q6_K_XL | 101.6 GB (3 parts) | 103.6 GB | Too big | Get file → | |
| Q8_0 | 117.5 GB (3 parts) | 119.5 GB | Too big | Get file → | |
| Q8_K_XL | 127.7 GB (3 parts) | 129.8 GB | Too big | Get file → | |
| BF16 | 221.0 GB (5 parts) | 223.1 GB | Too big | Get file → |
For 8 GB VRAM use Q4_0 (62.78 GB file, ~64.9 GB memory). For 16 GB VRAM use Q4_K_M (72.98 GB file, ~75.1 GB memory). For 32 GB VRAM use Q5_K_XL (82.85 GB file, ~84.9 GB memory). Estimates are approximate.
What GPU do you need for Glm-4.5-Air?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: the best quant that fits is Q2_K_L (45.5 GB file, about 47.5 GB in use).
Which Glm-4.5-Air quant should you choose?
Q4_K_M is the usual starting point for Glm-4.5-Air: it balances file size and quality well. Here it is a 73.0 GB file.
The Q8_0 file is 117.5 GB, about 1.6 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 TQ1_0 at 38.3 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Glm-4.5-Air
How much VRAM does Glm-4.5-Air need?
About 75.1 GB at Q4_K_M with an 8K context window. That is the 73.0 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 Glm-4.5-Air on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 40.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Glm-4.5-Air GGUF?
It depends on the quant: TQ1_0 is 38.3 GB, IQ1_S is 38.5 GB, IQ1_M is 40.1 GB, IQ2_XXS is 42.7 GB, IQ2_M is 44.4 GB, Q2_K is 45.3 GB and 21 more.
What context length does Glm-4.5-Air support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.3 GB at 32K.
Where can I download Glm-4.5-Air GGUF?
The files are on Hugging Face in the unsloth/GLM-4.5-Air-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 Glm-4.5-Air 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 GLM‑4.5‑Air on 8 GB VRAM?
Use the Q4_0 quant; it requires about 64.9 GB of VRAM at 8 k context, which is beyond 8 GB. A lower‑quality quant is not available for this model.
Does GLM‑4.5‑Air support vision?
No, the model has no vision or multimodal capabilities.