Granite 4.0 Tiny Base Preview needs about 6.6 GB of memory at Q4_K_M (a 4.1 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 16 quants, from 2.5 GB (Q2_K) to 13.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.
Granite 4.0 Tiny Base Preview is a 6.67B-parameter Granite model in GGUF format, quantized by mrutkows. Its trained context length is 1024K tokens. The repository has 16 quant files, from 2.5 GB (Q2_K) to 13.4 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 2.5 GB | 5.0 GB | Fits | Get file → | |
| Q3_K_S | 2.9 GB | 5.4 GB | Fits | Get file → | |
| Q3_K_M | 3.2 GB | 5.7 GB | Fits | Get file → | |
| Q3_K_L | 3.4 GB | 6.0 GB | Fits | Get file → | |
| Q4_0 | 3.8 GB | 6.3 GB | Fits | Get file → | |
| Q4_K_S | 3.8 GB | 6.3 GB | Fits | Get file → | |
| Q4_K_M | 4.1 GB | 6.6 GB | Fits | Get file → | |
| Q4_1 | 4.2 GB | 6.7 GB | Fits | Get file → | |
| Q5_0 | 4.6 GB | 7.1 GB | Fits | Get file → | |
| Q5_K_S | 4.6 GB | 7.1 GB | Fits | Get file → | |
| Q5_K_M | 4.7 GB | 7.3 GB | Tight | Get file → | |
| Q5_1 | 5.0 GB | 7.5 GB | Tight | Get file → | |
| Q6_K | 5.5 GB | 8.0 GB | Too big | Get file → | |
| Q8_0 | 7.1 GB | 9.6 GB | Too big | Get file → | |
| BF16 | 13.4 GB | 15.9 GB | Too big | Get file → | |
| F16 | 13.4 GB | 15.9 GB | Too big | Get file → |
What GPU do you need for Granite 4.0 Tiny Base Preview?
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_L (3.4 GB file, about 6.0 GB in use).
- 8 GB VRAM: the best quant that fits is Q6_K (5.5 GB file, about 8.0 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (7.1 GB file, about 9.6 GB in use).
- 16 GB VRAM: the best quant that fits is BF16 (13.4 GB file, about 15.9 GB in use).
Which Granite 4.0 Tiny Base Preview quant should you choose?
Q4_K_M is the usual starting point for Granite 4.0 Tiny Base Preview: it balances file size and quality well. Here it is a 4.1 GB file.
The Q8_0 file is 7.1 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 Q2_K at 2.5 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
Frequently asked questions about Granite 4.0 Tiny Base Preview
How much VRAM does Granite 4.0 Tiny Base Preview need?
About 6.6 GB at Q4_K_M with an 8K context window. That is the 4.1 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 4.0 Tiny Base Preview on an 8 GB GPU?
Yes, with Q6_K: the 5.5 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 4.0 Tiny Base Preview GGUF?
It depends on the quant: Q2_K is 2.5 GB, Q3_K_S is 2.9 GB, Q3_K_M is 3.2 GB, Q3_K_L is 3.4 GB, Q4_0 is 3.8 GB, Q4_K_S is 3.8 GB and 10 more.
What context length does Granite 4.0 Tiny Base Preview support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.
Where can I download Granite 4.0 Tiny Base Preview GGUF?
The files are on Hugging Face in the mrutkows/granite-4.0-tiny-base-preview-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 4.0 Tiny Base Preview 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.