GLM 4.5 Iceblink v3 106B A12B needs about 90.7 GB of memory at Q8_0 (a 88.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 65.9 GB (Q8_0) to 97.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.2 GB.
GLM 4.5 Iceblink v3 106B A12B – a large mixture‑of‑experts language model
GLM 4.5 Iceblink v3 106B A12B is a 107 billion‑parameter model with 12 billion active parameters. It supports a context length of 131 072 tokens and is released under the MIT license.
Good at
- Large context handling
- Mixture‑of‑experts inference
- High‑capacity language modeling
Limits
- Requires high VRAM for full precision
- No vision capabilities
- Large file sizes
Best for
- Long‑form content generation
- Complex reasoning tasks
- Research prototyping
Who it’s for. Developers and researchers needing a large, flexible language model with extensive context support. The model is suitable for environments that can accommodate its memory demands.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 65.9 GB (2 parts) | 67.9 GB | Too big | Get file → | |
| Q8_0 | 88.7 GB (3 parts) | 90.7 GB | Too big | Get file → | |
| Q8_0 | 97.5 GB (3 parts) | 99.6 GB | Too big | Get file → |
For 8 GB VRAM, use the Q8_0 quantized file of 65.86 GB; estimated memory is 67.9 GB at 8 k context. For 12 GB VRAM, the 88.7 GB Q8_0 file uses 90.7 GB memory. Larger VRAM allows the 97.52 GB file with 99.6 GB memory.
What GPU do you need for GLM 4.5 Iceblink v3 106B A12B?
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: no quant fits at an 8K context.
Which GLM 4.5 Iceblink v3 106B A12B quant should you choose?
Q8_0 (88.7 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is Q8_0 at 65.9 GB, with the quality trade-offs that come with fewer bits. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about GLM 4.5 Iceblink v3 106B A12B
How much VRAM does GLM 4.5 Iceblink v3 106B A12B need?
About 90.7 GB at Q8_0 with an 8K context window. That is the 88.7 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 GLM 4.5 Iceblink v3 106B A12B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 67.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of GLM 4.5 Iceblink v3 106B A12B GGUF?
It depends on the quant: Q8_0 is 65.9 GB, Q8_0 is 88.7 GB, Q8_0 is 97.5 GB.
What context length does GLM 4.5 Iceblink v3 106B A12B support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.5 GB at 8K and 6.2 GB at 32K.
Where can I download GLM 4.5 Iceblink v3 106B A12B GGUF?
The files are on Hugging Face in the zerofata/GLM-4.5-Iceblink-v3-106B-A12B-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 Iceblink v3 106B A12B 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 Iceblink v3 on 8 GB VRAM?
The smallest Q8_0 file requires 67.9 GB memory at 8 k context, so 8 GB VRAM is insufficient.
Does this model support vision tasks?
No, the model has no vision capabilities.
What license applies to this model?
The model is released under the MIT license.