StarCoder Models — Hardware Requirements
5 StarCoder models from BigCode and the community, from the smallest that runs in 0.5 GB of VRAM at Q4 up to 16.0B parameters. Every row links to full quantization tables and GPU compatibility.
All StarCoder Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Tiny Starcoder Py | 164M | 0.5 GB | 8K | ||
| Starcoder2 3B | 3.0B | 2.6 GB | 16K | ||
| Starcoder2 7B | 7.2B | 5.7 GB | 16K | ||
| Starcoder | 15.8B | 13.6 GB | — |
Community variants and fine-tunes(1)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Starcoder2 15B | 16.0B | 11.2 GB | 16K |
Frequently Asked Questions
- How much VRAM do I need to run a StarCoder model?
- The smallest StarCoder model, Tiny Starcoder Py, runs from 0.5 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.1 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which StarCoder models can I run on a 16 GB GPU?
- 5 of 5 StarCoder models fit in 16 GB of VRAM at Q4_K_M, including Starcoder2 3B, Starcoder2 15B, Starcoder2 7B.
- What is the most popular StarCoder model to run locally?
- Starcoder2 3B is the most downloaded StarCoder model in local-friendly quantized formats. It runs from 2.6 GB of VRAM at Q4_K_M.