GLM-4.5V needs about 65.7 GB of memory at Q4_K_M (a 63.6 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 4 quants, from 0.7 GB (Q4_0) to 113.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.2 GB.
GLM-4.5V – a 107B‑parameter multimodal LLM with reasoning, tools, and vision
GLM-4.5V is a 107B‑parameter model in the GLM family, built on the glm4moe architecture. It supports text, image, video, and document understanding, and offers a Thinking Mode for deeper reasoning. The model is released under the MIT license.
Good at
- Multimodal reasoning
- Vision‑language tasks
- Tool usage
Limits
- Requires large VRAM for full model
- No native vision support in llama.cpp
- Limited to 65536 context tokens
Best for
- Complex document analysis
- GUI automation
- Long‑form multimodal dialogue
Who it’s for. Developers needing a versatile multimodal model for research or production. The model is open‑source and can be run locally with llama.cpp or used via the provided demo and API.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 0.7 GB | 2.8 GB | Fits | Get file → | |
| Q8_0 | 1.3 GB | 3.4 GB | Fits | Get file → | |
| Q4_K_M | 63.6 GB (2 parts) | 65.7 GB | Too big | Get file → | |
| Q8_0 | 113.6 GB (2 parts) | 115.7 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (0.71 GB, 2.8 GB memory at 8 k context). For 16 GB VRAM, Q8_0 (1.33 GB, 3.4 GB memory). Larger models (Q4_K_M, Q8_0 shards) require 64 GB+ VRAM and are not recommended for typical setups.
What GPU do you need for GLM-4.5V?
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 Q8_0 (1.3 GB file, about 3.4 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (1.3 GB file, about 3.4 GB in use).
Which GLM-4.5V quant should you choose?
Q4_K_M is the usual starting point for GLM-4.5V: it balances file size and quality well. Here it is a 63.6 GB file.
The Q8_0 file is 1.3 GB, about 0.0 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 Q4_0 at 0.7 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about GLM-4.5V
How much VRAM does GLM-4.5V need?
About 65.7 GB at Q4_K_M with an 8K context window. That is the 63.6 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-4.5V on an 8 GB GPU?
Yes, with Q8_0: the 1.3 GB file needs about 3.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of GLM-4.5V GGUF?
It depends on the quant: Q4_0 is 0.7 GB, Q8_0 is 1.3 GB, Q4_K_M is 63.6 GB, Q8_0 is 113.6 GB.
What context length does GLM-4.5V support?
The GGUF header lists a trained context length of 65,536 tokens (64K). 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-4.5V GGUF?
The files are on Hugging Face in the ggml-org/GLM-4.5V-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-4.5V 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-4.5V on 8 GB VRAM?
Yes, using the Q4_0 quantized file (0.71 GB) which requires about 2.8 GB VRAM at 8 k context.
Does the model support vision tasks in llama.cpp?
No, llama.cpp does not natively support vision; the model is intended for use with the provided demo or API.
What is the maximum context length?
The model was trained on 65536 context tokens.