EXAONE-4.5-33B needs about 22.7 GB of memory at Q4_K_M (a 20.0 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 6 quants, from 18.0 GB (IQ4_XS) to 66.1 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.7 GB.
EXAONE‑4.5‑33B – 33B parameter vision‑language model
EXAONE‑4.5‑33B is a 33‑billion‑parameter vision‑language model from LG AI Research. It combines a 31.7B language backbone with a 1.29B vision encoder. The model supports up to 262,144 tokens of context and was trained with a Dec 2024 knowledge cutoff.
Good at
- Vision‑language reasoning
- Document understanding
- Korean contextual reasoning
Limits
- No vision support in quantized GGUF files
- Large VRAM requirement for full precision
- License type is other, not open‑source
Best for
- Multimodal document analysis
- Korean language content generation
- Vision‑based QA tasks
Who it’s for. Developers building multimodal applications that need strong language and vision capabilities. Researchers interested in open‑weight vision‑language models with large context windows.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ4_XS | 18.0 GB | 20.7 GB | Too big | Get file → | |
| Q4_K_M | 20.0 GB | 22.7 GB | Too big | Get file → | |
| Q5_K_M | 23.5 GB | 26.2 GB | Too big | Get file → | |
| Q6_K | 27.1 GB | 29.8 GB | Too big | Get file → | |
| Q8_0 | 35.1 GB | 37.8 GB | Too big | Get file → | |
| BF16 | 66.1 GB | 68.8 GB | Too big | Get file → |
For 8 GB VRAM use IQ4_XS (18.03 GB file, ~20.7 GB memory at 8k context). For 16 GB VRAM use Q4_K_M (20.05 GB file, ~22.7 GB memory). For 32 GB VRAM use Q6_K (27.13 GB file, ~29.8 GB memory).
What GPU do you need for EXAONE-4.5-33B?
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: the best quant that fits is Q4_K_M (20.0 GB file, about 22.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (27.1 GB file, about 29.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (35.1 GB file, about 37.8 GB in use).
Which EXAONE-4.5-33B quant should you choose?
Q4_K_M is the usual starting point for EXAONE-4.5-33B: it balances file size and quality well. Here it is a 20.0 GB file.
The Q8_0 file is 35.1 GB, about 1.8 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 IQ4_XS at 18.0 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 EXAONE-4.5-33B
How much VRAM does EXAONE-4.5-33B need?
About 22.7 GB at Q4_K_M with an 8K context window. That is the 20.0 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 EXAONE-4.5-33B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 20.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of EXAONE-4.5-33B GGUF?
It depends on the quant: IQ4_XS is 18.0 GB, Q4_K_M is 20.0 GB, Q5_K_M is 23.5 GB, Q6_K is 27.1 GB, Q8_0 is 35.1 GB, BF16 is 66.1 GB.
What context length does EXAONE-4.5-33B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.2 GB at 8K and 8.7 GB at 32K.
Where can I download EXAONE-4.5-33B GGUF?
The files are on Hugging Face in the LGAI-EXAONE/EXAONE-4.5-33B-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 EXAONE-4.5-33B use?
Hugging Face lists the license as other. 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 EXAONE‑4.5‑33B on 8 GB VRAM?
The smallest quantized file is IQ4_XS, which requires about 20.7 GB memory at 8k context, so 8 GB VRAM is insufficient.
Does the model support vision tasks?
Yes, it includes a dedicated vision encoder and supports vision‑language reasoning.
What is the maximum context length?
The model supports up to 262,144 tokens of context.