granite-20b-code-base-8k needs about 13.4 GB of memory at Q4_K_M (a 12.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 29 quants, from 6.1 GB (IQ1_S) to 40.2 GB (BF16).
Granite‑20B‑Code‑Base‑8K – a 20.1B parameter code generation model
Granite‑20B‑Code‑Base‑8K is a decoder‑only model trained on 3 trillion code tokens and 500 billion mixed code and natural language tokens. It supports 8 k context length and is released under Apache‑2.0.
Good at
- Code generation
- Code explanation
- Code fixing
Limits
- No vision support
- Requires 8 k context
- Large VRAM for higher quants
Best for
- Generating boilerplate code
- Explaining code snippets
- Fixing bugs in source files
Who it’s for. Developers and teams needing a dedicated code model for generation and analysis tasks. The model is open source and can be run locally with llama.cpp.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 6.1 GB | 6.8 GB | Fits | Get file → | |
| IQ1_M | 6.5 GB | 7.3 GB | Tight | Get file → | |
| IQ2_XXS | 6.6 GB | 7.3 GB | Tight | Get file → | |
| IQ2_XS | 7.1 GB | 7.8 GB | Tight | Get file → | |
| IQ2_S | 7.5 GB | 8.2 GB | Too big | Get file → | |
| IQ2_M | 7.7 GB | 8.4 GB | Too big | Get file → | |
| Q2_K_S | 8.0 GB | 8.7 GB | Too big | Get file → | |
| Q2_K_M | 8.1 GB | 8.8 GB | Too big | Get file → | |
| IQ3_XXS | 8.8 GB | 9.5 GB | Too big | Get file → | |
| IQ3_XS | 8.9 GB | 9.6 GB | Too big | Get file → | |
| IQ3_M | 10.0 GB | 10.7 GB | Too big | Get file → | |
| IQ3_S | 10.0 GB | 10.7 GB | Too big | Get file → | |
| Q3_K_S | 10.4 GB | 11.1 GB | Too big | Get file → | |
| Q3_K_M | 10.5 GB | 11.2 GB | Too big | Get file → | |
| IQ4_XS | 10.9 GB | 11.7 GB | Too big | Get file → | |
| Q4_0 | 11.5 GB | 12.2 GB | Too big | Get file → | |
| IQ4_NL | 11.6 GB | 12.3 GB | Too big | Get file → | |
| Q4_K_S | 11.9 GB | 12.6 GB | Too big | Get file → | |
| Q4_K_M | 12.6 GB | 13.4 GB | Too big | Get file → | |
| Q4_1 | 12.7 GB | 13.4 GB | Too big | Get file → | |
| Q5_0 | 14.0 GB | 14.7 GB | Too big | Get file → | |
| Q5_K_S | 14.4 GB | 15.1 GB | Too big | Get file → | |
| Q5_K_M | 14.9 GB | 15.6 GB | Too big | Get file → | |
| Q5_1 | 15.2 GB | 15.9 GB | Too big | Get file → | |
| Q6_K_M | 16.6 GB | 17.4 GB | Too big | Get file → | |
| Q8_0 | 21.5 GB | 22.2 GB | Too big | Get file → | |
| BF16 | 32.9 GB | 33.6 GB | Too big | Get file → | |
| F16 | 32.9 GB | 33.6 GB | Too big | Get file → | |
| BF16 | 40.2 GB | 41.0 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (6.07 GB) or IQ1_M (6.54 GB). For 12 GB VRAM, IQ4_0 (11.47 GB) or Q4_K_S (11.89 GB) are suitable. Higher quants like Q5_K_M (14.87 GB) need 16 GB or more.
What GPU do you need for granite-20b-code-base-8k?
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: the best quant that fits is IQ2_XS (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_1 (15.2 GB file, about 15.9 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 BF16 (40.2 GB file, about 41.0 GB in use).
Which granite-20b-code-base-8k quant should you choose?
Q4_K_M is the usual starting point for granite-20b-code-base-8k: it balances file size and quality well. Here it is a 12.6 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 6.1 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-base-8k
How much VRAM does granite-20b-code-base-8k need?
About 13.4 GB at Q4_K_M with an 8K context window. That is the 12.6 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-base-8k on an 8 GB GPU?
Yes, with IQ2_XS: 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-base-8k GGUF?
It depends on the quant: IQ1_S is 6.1 GB, IQ1_M is 6.5 GB, IQ2_XXS is 6.6 GB, IQ2_XS is 7.1 GB, IQ2_S is 7.5 GB, IQ2_M is 7.7 GB and 23 more.
What context length does granite-20b-code-base-8k 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-base-8k GGUF?
The files are on Hugging Face in the Mungert/granite-20b-code-base-8k-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-base-8k 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 this model on 8 GB VRAM?
Yes, use IQ1_S or IQ1_M quantizations, which fit within 8 GB VRAM.
Does this model support image inputs?
No, it is a text‑only code model with no vision capability.
What license does it use?
Apache‑2.0, allowing free use and modification.