granite-3.3-8b-instruct needs about 6.8 GB of memory at Q4_K_M (a 4.9 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 25 quants, from 2.3 GB (IQ2_XXS) to 16.3 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 5.4 GB.
Granite‑3.3‑8B‑Instruct – 8.17B parameter instruction model
Granite‑3.3‑8B‑Instruct is a decoder‑only language model with a 131072 token context window. It is trained on a large corpus and supports reasoning and tool use. The model is released under Apache‑2.0.
Good at
- Reasoning
- Tool use
- Long‑context text generation
Limits
- No vision support
- Large memory requirement for full context
- No mixture of experts
Best for
- Code completion with FIM
- Long‑form summarization
- Question answering
Who it’s for. Developers and researchers needing a large instruction‑tuned model with a wide context window. The model is suitable for applications that require reasoning and tool integration.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 2.3 GB | 4.1 GB | Fits | Get file → | |
| IQ2_M | 2.8 GB | 4.7 GB | Fits | Get file → | |
| Q2_K_S | 2.9 GB | 4.7 GB | Fits | Get file → | |
| Q2_K_M | 3.1 GB | 4.9 GB | Fits | Get file → | |
| IQ3_XXS | 3.2 GB | 5.0 GB | Fits | Get file → | |
| IQ3_XS | 3.4 GB | 5.2 GB | Fits | Get file → | |
| Q3_K_S | 3.6 GB | 5.4 GB | Fits | Get file → | |
| IQ3_S | 3.6 GB | 5.4 GB | Fits | Get file → | |
| IQ3_M | 3.7 GB | 5.6 GB | Fits | Get file → | |
| Q3_K_M | 4.0 GB | 5.8 GB | Fits | Get file → | |
| IQ4_XS | 4.4 GB | 6.3 GB | Fits | Get file → | |
| Q4_0 | 4.6 GB | 6.4 GB | Fits | Get file → | |
| IQ4_NL | 4.7 GB | 6.5 GB | Fits | Get file → | |
| Q4_K_S | 4.7 GB | 6.5 GB | Fits | Get file → | |
| Q4_K_M | 4.9 GB | 6.8 GB | Fits | Get file → | |
| Q4_1 | 5.1 GB | 7.0 GB | Fits | Get file → | |
| Q5_0 | 5.6 GB | 7.5 GB | Tight | Get file → | |
| Q5_K_S | 5.6 GB | 7.5 GB | Tight | Get file → | |
| Q5_K_M | 5.8 GB | 7.6 GB | Tight | Get file → | |
| Q5_1 | 6.1 GB | 8.0 GB | Tight | Get file → | |
| Q6_K_M | 6.7 GB | 8.5 GB | Too big | Get file → | |
| Q8_0 | 8.7 GB | 10.5 GB | Too big | Get file → | |
| BF16 | 8.9 GB | 10.7 GB | Too big | Get file → | |
| F16 | 8.9 GB | 10.7 GB | Too big | Get file → | |
| BF16 | 16.3 GB | 18.2 GB | Too big | Get file → |
For 8 GB VRAM, Q4_K_M (file size 4.94 GB, estimated memory 6.8 GB at 8 k context) is a good balance of size and performance. If more VRAM is available, Q5_K_M (5.8 GB, 7.6 GB) offers higher precision. For very limited memory, IQ2_XXS (2.27 GB, 4.1 GB) can be used.
What GPU do you need for granite-3.3-8b-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: the best quant that fits is Q3_K_M (4.0 GB file, about 5.8 GB in use).
- 8 GB VRAM: the best quant that fits is Q5_1 (6.1 GB file, about 8.0 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (8.9 GB file, about 10.7 GB in use).
- 16 GB VRAM: the best quant that fits is BF16 (8.9 GB file, about 10.7 GB in use).
- 24 GB VRAM: the best quant that fits is BF16 (16.3 GB file, about 18.2 GB in use).
Which granite-3.3-8b-instruct quant should you choose?
Q4_K_M is the usual starting point for granite-3.3-8b-instruct: it balances file size and quality well. Here it is a 4.9 GB file.
The Q8_0 file is 8.7 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 2.3 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 granite-3.3-8b-instruct
How much VRAM does granite-3.3-8b-instruct need?
About 6.8 GB at Q4_K_M with an 8K context window. That is the 4.9 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 granite-3.3-8b-instruct on an 8 GB GPU?
Yes, with Q5_1: the 6.1 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of granite-3.3-8b-instruct GGUF?
It depends on the quant: IQ2_XXS is 2.3 GB, IQ2_M is 2.8 GB, Q2_K_S is 2.9 GB, Q2_K_M is 3.1 GB, IQ3_XXS is 3.2 GB, IQ3_XS is 3.4 GB and 19 more.
What context length does granite-3.3-8b-instruct support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.3 GB at 8K and 5.4 GB at 32K.
Where can I download granite-3.3-8b-instruct GGUF?
The files are on Hugging Face in the Mungert/granite-3.3-8b-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 granite-3.3-8b-instruct use?
Hugging Face lists the license as apache-2.0. 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 Q4_K_M on 8 GB VRAM?
Yes, Q4_K_M requires about 6.8 GB of VRAM at 8 k context, which fits within an 8 GB budget.
Is there a quantization that fits 4 GB VRAM?
IQ2_XXS uses 2.27 GB of storage and needs about 4.1 GB of VRAM at 8 k context, suitable for 4 GB setups.
Does this model support vision tasks?
No, the model has no vision capabilities and is text‑only.