granite-3b-code-base-128k needs about 5.3 GB of memory at Q4_K_M (a 2.1 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 27 quants, from 1.1 GB (IQ2_XXS) to 7.0 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Granite-3B-Code-Base-128K: 3.48B Llama model with 128K token context for code tasks
Granite-3B-Code-Base-128K is a 3.48B parameter Llama model trained for code generation and related tasks. It supports a 128000 token context window and is released under Apache-2.0. The model is available in multiple GGUF quantizations.
Good at
- Long‑context code generation
- Code explanation
- Unit test generation
Limits
- No vision support
- Requires 128K context for full capability
- License is Apache-2.0
Best for
- Enterprise code productivity tools
- Large‑scale code review assistants
- Code‑centric chatbots
Who it’s for. Developers and enterprises needing deep code understanding and generation over long documents. The model is suitable for teams building code‑centric AI services.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 1.1 GB | 4.3 GB | Fits | Get file → | |
| IQ2_XS | 1.2 GB | 4.4 GB | Fits | Get file → | |
| IQ2_S | 1.2 GB | 4.4 GB | Fits | Get file → | |
| IQ2_M | 1.3 GB | 4.5 GB | Fits | Get file → | |
| Q2_K_S | 1.3 GB | 4.5 GB | Fits | Get file → | |
| Q2_K_M | 1.3 GB | 4.5 GB | Fits | Get file → | |
| IQ3_XXS | 1.4 GB | 4.6 GB | Fits | Get file → | |
| IQ3_XS | 1.5 GB | 4.7 GB | Fits | Get file → | |
| IQ3_M | 1.7 GB | 4.8 GB | Fits | Get file → | |
| IQ3_S | 1.7 GB | 4.8 GB | Fits | Get file → | |
| Q3_K_S | 1.7 GB | 4.9 GB | Fits | Get file → | |
| Q3_K_M | 1.7 GB | 4.9 GB | Fits | Get file → | |
| IQ4_XS | 1.9 GB | 5.1 GB | Fits | Get file → | |
| Q4_0 | 2.0 GB | 5.1 GB | Fits | Get file → | |
| IQ4_NL | 2.0 GB | 5.2 GB | Fits | Get file → | |
| Q4_K_S | 2.0 GB | 5.2 GB | Fits | Get file → | |
| Q4_K_M | 2.1 GB | 5.3 GB | Fits | Get file → | |
| Q4_1 | 2.2 GB | 5.4 GB | Fits | Get file → | |
| Q5_0 | 2.4 GB | 5.6 GB | Fits | Get file → | |
| Q5_K_S | 2.5 GB | 5.6 GB | Fits | Get file → | |
| Q5_K_M | 2.5 GB | 5.7 GB | Fits | Get file → | |
| Q5_1 | 2.6 GB | 5.8 GB | Fits | Get file → | |
| Q6_K_M | 2.9 GB | 6.0 GB | Fits | Get file → | |
| Q8_0 | 3.7 GB | 6.9 GB | Fits | Get file → | |
| BF16 | 5.0 GB | 8.2 GB | Too big | Get file → | |
| F16 | 5.0 GB | 8.2 GB | Too big | Get file → | |
| BF16 | 7.0 GB | 10.2 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (file size 1.96 GB, estimated 5.1 GB memory). For 12 GB VRAM, Q5_0 (2.4 GB, 5.6 GB memory) or Q5_K_M (2.49 GB, 5.7 GB memory). For 16 GB VRAM, Q6_K_M (2.86 GB, 6 GB memory) or Q8_0 (3.71 GB, 6.9 GB memory).
What GPU do you need for granite-3b-code-base-128k?
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 Q6_K_M (2.9 GB file, about 6.0 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (3.7 GB file, about 6.9 GB in use).
- 12 GB VRAM: the best quant that fits is BF16 (7.0 GB file, about 10.2 GB in use).
Which granite-3b-code-base-128k quant should you choose?
Q4_K_M is the usual starting point for granite-3b-code-base-128k: it balances file size and quality well. Here it is a 2.1 GB file.
The Q8_0 file is 3.7 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 1.1 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-3b-code-base-128k
How much VRAM does granite-3b-code-base-128k need?
About 5.3 GB at Q4_K_M with an 8K context window. That is the 2.1 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-3b-code-base-128k on an 8 GB GPU?
Yes, with Q8_0: the 3.7 GB file needs about 6.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of granite-3b-code-base-128k GGUF?
It depends on the quant: IQ2_XXS is 1.1 GB, IQ2_XS is 1.2 GB, IQ2_S is 1.2 GB, IQ2_M is 1.3 GB, Q2_K_S is 1.3 GB, Q2_K_M is 1.3 GB and 21 more.
What context length does granite-3b-code-base-128k support?
The GGUF header lists a trained context length of 128,000 tokens (125K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.
Where can I download granite-3b-code-base-128k GGUF?
The files are on Hugging Face in the Mungert/granite-3b-code-base-128k-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-3b-code-base-128k 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, using the Q4_0 quantization which requires about 5.1 GB of VRAM at 8K context.
Does the model support image input?
No, the model has no vision capabilities.
What license applies to this model?
The model is released under the Apache-2.0 license.