Granite 20b Code Instruct needs about 13.5 GB of memory at Q4_K_M (a 12.8 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 33 quants, from 4.5 GB (IQ1_S) to 80.3 GB (F32).
Granite 20b Code Instruct – 20B parameter code‑focused model
Granite 20b Code Instruct is a 20B parameter model fine‑tuned from Granite‑20B‑Code‑Base‑8K for instruction following and coding tasks. It supports 8k token context and is released under Apache 2.0. The model is available in multiple GGUF quantizations.
Good at
- Instruction following for code
- Logical reasoning in programming
- Tool‑aware code generation
Limits
- No vision support
- Requires 8k context for full capability
- Not recommended for new projects
Best for
- Building coding assistants
- Generating code snippets
- Answering programming questions
Who it’s for. Developers and researchers needing a code‑centric language model. It is suitable for projects that can accommodate its 8k context and 20B parameter size, and for those who prefer an Apache 2.0 licensed model.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 4.5 GB | 5.2 GB | Fits | Get file → | |
| IQ1_M | 4.9 GB | 5.6 GB | Fits | Get file → | |
| IQ2_XXS | 5.6 GB | 6.3 GB | Fits | Get file → | |
| IQ2_XS | 6.2 GB | 6.9 GB | Fits | Get file → | |
| IQ2_S | 6.5 GB | 7.2 GB | Tight | Get file → | |
| IQ2_M | 7.1 GB | 7.8 GB | Tight | Get file → | |
| Q2_K | 7.9 GB | 8.6 GB | Too big | Get file → | |
| Q2_K_L | 8.0 GB | 8.7 GB | Too big | Get file → | |
| IQ3_XXS | 8.1 GB | 8.8 GB | Too big | Get file → | |
| IQ3_XS | 8.7 GB | 9.4 GB | Too big | Get file → | |
| IQ3_S | 8.9 GB | 9.7 GB | Too big | Get file → | |
| Q3_K_S | 8.9 GB | 9.7 GB | Too big | Get file → | |
| IQ3_M | 9.6 GB | 10.3 GB | Too big | Get file → | |
| Q3_K_M | 10.6 GB | 11.3 GB | Too big | Get file → | |
| IQ4_XS | 10.9 GB | 11.7 GB | Too big | Get file → | |
| IQ4_NL | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_0 | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_0 | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_0 | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_0 | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_K_S | 11.7 GB | 12.4 GB | Too big | Get file → | |
| Q3_K_L | 11.7 GB | 12.5 GB | Too big | Get file → | |
| Q3_K_XL | 11.8 GB | 12.5 GB | Too big | Get file → | |
| Q4_K_M | 12.8 GB | 13.5 GB | Too big | Get file → | |
| Q4_K_L | 12.9 GB | 13.6 GB | Too big | Get file → | |
| Q5_K_S | 14.0 GB | 14.7 GB | Too big | Get file → | |
| Q5_K_M | 14.8 GB | 15.5 GB | Too big | Get file → | |
| Q5_K_L | 14.9 GB | 15.6 GB | Too big | Get file → | |
| Q6_K | 16.6 GB | 17.4 GB | Too big | Get file → | |
| Q6_K_L | 16.7 GB | 17.4 GB | Too big | Get file → | |
| Q8_0 | 21.5 GB | 22.2 GB | Too big | Get file → | |
| F16 | 40.2 GB | 41.0 GB | Too big | Get file → | |
| F32 | 80.3 GB (3 parts) | 81.0 GB | Too big | Get file → |
For 8 GB VRAM use IQ1_S (4.52 GB) or IQ1_M (4.91 GB). For 12 GB VRAM use Q4_0 (11.55 GB) or Q4_K_S (11.67 GB). For 16 GB VRAM use Q5_K_M (14.81 GB) or Q6_K (16.63 GB). Estimates are approximate.
What GPU do you need for Granite 20b Code 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 IQ1_M (4.9 GB file, about 5.6 GB in use).
- 8 GB VRAM: the best quant that fits is IQ2_M (7.1 GB file, about 7.8 GB in use).
- 12 GB VRAM: the best quant that fits is IQ4_XS (10.9 GB file, about 11.7 GB in use).
- 16 GB VRAM: the best quant that fits is Q5_K_L (14.9 GB file, about 15.6 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (21.5 GB file, about 22.2 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (21.5 GB file, about 22.2 GB in use).
- 48 GB VRAM: the best quant that fits is F16 (40.2 GB file, about 41.0 GB in use).
Which Granite 20b Code Instruct quant should you choose?
Q4_K_M is the usual starting point for Granite 20b Code Instruct: it balances file size and quality well. Here it is a 12.8 GB file.
The Q8_0 file is 21.5 GB, about 1.7 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 IQ1_S at 4.5 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
Source
Frequently asked questions about Granite 20b Code Instruct
How much VRAM does Granite 20b Code Instruct need?
About 13.5 GB at Q4_K_M with an 8K context window. That is the 12.8 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 20b Code Instruct on an 8 GB GPU?
Yes, with IQ2_M: the 7.1 GB file needs about 7.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Granite 20b Code Instruct GGUF?
It depends on the quant: IQ1_S is 4.5 GB, IQ1_M is 4.9 GB, IQ2_XXS is 5.6 GB, IQ2_XS is 6.2 GB, IQ2_S is 6.5 GB, IQ2_M is 7.1 GB and 27 more.
What context length does Granite 20b Code Instruct support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.2 GB at 8K.
Where can I download Granite 20b Code Instruct GGUF?
The files are on Hugging Face in the bartowski/granite-20b-code-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 20b Code 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 Granite 20b Code Instruct on 8 GB VRAM?
Yes, using the IQ1_S or IQ1_M quantizations which are around 4.5–5 GB in size and require about 5–6 GB of VRAM at 8k context.
Does this model support image inputs?
No, the model does not have vision capabilities.
What license applies to this model?
The model is released under the Apache 2.0 license.