Starcoder2 15b needs about 11.0 GB of memory at Q4_K_M (a 9.9 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 24 quants, from 3.6 GB (IQ1_S) to 13.1 GB (Q6_K).
Starcoder2 15b – 15B‑parameter coding model with 16,384‑token context
Starcoder2 15b is a 15B‑parameter transformer decoder trained on 600+ programming languages. It uses Grouped Query Attention and a sliding window of 4,096 tokens. The model is not an instruction model and is best used for code generation tasks.
Good at
- Code generation in many languages
- Fill‑in‑the‑Middle generation
- Large context handling
Limits
- Not an instruction model
- No vision support
- Requires large VRAM for full precision
Best for
- Generating code snippets
- Fine‑tuning on domain code
- Large‑context code analysis
Who it’s for. Developers and researchers needing a large coding model for code generation and analysis. The model is suitable for fine‑tuning on specialized codebases.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 3.6 GB | 4.7 GB | Fits | Get file → | |
| IQ1_M | 3.9 GB | 5.0 GB | Fits | Get file → | |
| IQ2_XXS | 4.4 GB | 5.5 GB | Fits | Get file → | |
| IQ2_XS | 4.8 GB | 6.0 GB | Fits | Get file → | |
| IQ2_S | 5.1 GB | 6.3 GB | Fits | Get file → | |
| IQ2_M | 5.5 GB | 6.7 GB | Fits | Get file → | |
| Q2_K | 6.2 GB | 7.4 GB | Tight | Get file → | |
| IQ3_XXS | 6.2 GB | 7.4 GB | Tight | Get file → | |
| IQ3_XS | 6.7 GB | 7.9 GB | Tight | Get file → | |
| Q3_K_S | 7.0 GB | 8.2 GB | Too big | Get file → | |
| IQ3_S | 7.0 GB | 8.2 GB | Too big | Get file → | |
| IQ3_M | 7.3 GB | 8.5 GB | Too big | Get file → | |
| Q3_K_M | 8.0 GB | 9.2 GB | Too big | Get file → | |
| IQ4_XS | 8.6 GB | 9.8 GB | Too big | Get file → | |
| Q3_K_L | 9.0 GB | 10.1 GB | Too big | Get file → | |
| Q4_0 | 9.1 GB | 10.2 GB | Too big | Get file → | |
| Q4_0 | 9.1 GB | 10.2 GB | Too big | Get file → | |
| Q4_0 | 9.1 GB | 10.2 GB | Too big | Get file → | |
| Q4_0 | 9.1 GB | 10.3 GB | Too big | Get file → | |
| Q4_K_S | 9.2 GB | 10.3 GB | Too big | Get file → | |
| Q4_K_M | 9.9 GB | 11.0 GB | Too big | Get file → | |
| Q5_K_S | 11.0 GB | 12.2 GB | Too big | Get file → | |
| Q5_K_M | 11.4 GB | 12.6 GB | Too big | Get file → | |
| Q6_K | 13.1 GB | 14.3 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (3.56 GB, 4.7 GB memory at 8k context). For 12 GB VRAM, IQ2_M (5.55 GB, 6.7 GB memory). For 16 GB VRAM, Q4_K_M (9.86 GB, 11 GB memory). Estimates are approximate.
What GPU do you need for Starcoder2 15b?
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 IQ2_XS (4.8 GB file, about 6.0 GB in use).
- 8 GB VRAM: the best quant that fits is IQ3_XS (6.7 GB file, about 7.9 GB in use).
- 12 GB VRAM: the best quant that fits is Q4_K_M (9.9 GB file, about 11.0 GB in use).
- 16 GB VRAM: the best quant that fits is Q6_K (13.1 GB file, about 14.3 GB in use).
Which Starcoder2 15b quant should you choose?
Q4_K_M is the usual starting point for Starcoder2 15b: it balances file size and quality well. Here it is a 9.9 GB file.
If memory is tight, the smallest file is IQ1_S at 3.6 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 Starcoder2 15b
How much VRAM does Starcoder2 15b need?
About 11.0 GB at Q4_K_M with an 8K context window. That is the 9.9 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 Starcoder2 15b on an 8 GB GPU?
Yes, with IQ3_XS: the 6.7 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Starcoder2 15b GGUF?
It depends on the quant: IQ1_S is 3.6 GB, IQ1_M is 3.9 GB, IQ2_XXS is 4.4 GB, IQ2_XS is 4.8 GB, IQ2_S is 5.1 GB, IQ2_M is 5.5 GB and 18 more.
What context length does Starcoder2 15b support?
The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 0.7 GB at 8K and 1.3 GB at 16K.
Where can I download Starcoder2 15b GGUF?
The files are on Hugging Face in the mradermacher/starcoder2-15b-i1-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 Starcoder2 15b use?
Hugging Face lists the license as bigcode-openrail-m. 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 Starcoder2 15b on 8 GB VRAM?
Use the IQ1_S quantized file; it requires about 4.7 GB VRAM at 8k context.
Does the model support image inputs?
No, Starcoder2 15b does not have vision capabilities.
Is the model licensed for commercial use?
The model is released under bigcode-openrail-m, which allows commercial use with attribution.