EXAONE 3.5 32B Instruct needs about 22.0 GB of memory at Q4_K_M (a 19.3 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 28 quants, from 8.7 GB (IQ2_XXS) to 64.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.
EXAONE 3.5 32B Instruct – 32B instruction‑tuned model with 32K context
EXAONE 3.5 32B Instruct is a 32B parameter instruction‑tuned model from LG AI Research. It supports up to 32,768 token context and is available in multiple GGUF quantizations. The model is designed for long‑context understanding and bilingual instruction handling.
Good at
- Long‑context instruction following
- Bilingual (English/Korean) generation
- Large‑scale conversational tasks
Limits
- No vision support
- Requires 8,000 token context for memory estimates
- License is other, not specified
Best for
- Chatbot development
- Document summarization
- Code generation
Who it’s for. Developers needing a large instruction‑tuned model for text‑only applications. Suitable for researchers and enterprises that can host the model locally or in the cloud.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 8.7 GB | 11.3 GB | Too big | Get file → | |
| IQ2_XS | 9.6 GB | 12.3 GB | Too big | Get file → | |
| IQ2_S | 10.0 GB | 12.7 GB | Too big | Get file → | |
| IQ2_M | 10.9 GB | 13.5 GB | Too big | Get file → | |
| Q2_K | 11.9 GB | 14.6 GB | Too big | Get file → | |
| Q2_K_L | 12.4 GB | 15.1 GB | Too big | Get file → | |
| IQ3_XS | 13.3 GB | 15.9 GB | Too big | Get file → | |
| Q3_K_S | 14.0 GB | 16.6 GB | Too big | Get file → | |
| IQ3_M | 14.4 GB | 17.0 GB | Too big | Get file → | |
| Q3_K_M | 15.5 GB | 18.1 GB | Too big | Get file → | |
| Q3_K_L | 16.8 GB | 19.4 GB | Too big | Get file → | |
| IQ4_XS | 17.2 GB | 19.9 GB | Too big | Get file → | |
| Q3_K_XL | 17.3 GB | 19.9 GB | Too big | Get file → | |
| Q4_0 | 18.1 GB | 20.8 GB | Too big | Get file → | |
| Q4_0 | 18.1 GB | 20.8 GB | Too big | Get file → | |
| Q4_0 | 18.1 GB | 20.8 GB | Too big | Get file → | |
| IQ4_NL | 18.2 GB | 20.8 GB | Too big | Get file → | |
| Q4_0 | 18.2 GB | 20.9 GB | Too big | Get file → | |
| Q4_K_S | 18.3 GB | 20.9 GB | Too big | Get file → | |
| Q4_K_M | 19.3 GB | 22.0 GB | Too big | Get file → | |
| Q4_K_L | 19.7 GB | 22.4 GB | Too big | Get file → | |
| Q5_K_S | 22.1 GB | 24.7 GB | Too big | Get file → | |
| Q5_K_M | 22.7 GB | 25.3 GB | Too big | Get file → | |
| Q5_K_L | 23.0 GB | 25.7 GB | Too big | Get file → | |
| Q6_K | 26.3 GB | 28.9 GB | Too big | Get file → | |
| Q6_K_L | 26.5 GB | 29.2 GB | Too big | Get file → | |
| Q8_0 | 34.0 GB | 36.7 GB | Too big | Get file → | |
| F16 | 64.0 GB (2 parts) | 66.7 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ2_XXS (8.7 GB file, 11.3 GB memory at 8K context). For 16‑GB VRAM, use Q4_K_M (19.34 GB file, 22 GB memory). For 32‑GB VRAM, use Q6_K (26.26 GB file, 28.9 GB memory). Estimates are approximate.
What GPU do you need for EXAONE 3.5 32B Instruct?
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: the best quant that fits is IQ2_XXS (8.7 GB file, about 11.3 GB in use).
- 16 GB VRAM: the best quant that fits is IQ3_XS (13.3 GB file, about 15.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q4_K_L (19.7 GB file, about 22.4 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K_L (26.5 GB file, about 29.2 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (34.0 GB file, about 36.7 GB in use).
Which EXAONE 3.5 32B Instruct quant should you choose?
Q4_K_M is the usual starting point for EXAONE 3.5 32B Instruct: it balances file size and quality well. Here it is a 19.3 GB file.
The Q8_0 file is 34.0 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 IQ2_XXS at 8.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about EXAONE 3.5 32B Instruct
How much VRAM does EXAONE 3.5 32B Instruct need?
About 22.0 GB at Q4_K_M with an 8K context window. That is the 19.3 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 3.5 32B Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 11.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of EXAONE 3.5 32B Instruct GGUF?
It depends on the quant: IQ2_XXS is 8.7 GB, IQ2_XS is 9.6 GB, IQ2_S is 10.0 GB, IQ2_M is 10.9 GB, Q2_K is 11.9 GB, Q2_K_L is 12.4 GB and 22 more.
What context length does EXAONE 3.5 32B Instruct support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.
Where can I download EXAONE 3.5 32B Instruct GGUF?
The files are on Hugging Face in the bartowski/EXAONE-3.5-32B-Instruct-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 3.5 32B Instruct 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 3.5 32B Instruct on 8 GB VRAM?
Use the IQ2_XXS quant, which has a file size of 8.7 GB and requires about 11.3 GB of VRAM at 8K context.
Does the model support vision tasks?
No, the model is text‑only and does not include vision capabilities.
What is the maximum context length?
The model supports up to 32,768 tokens of context.