BigCode
StarCoder2 15B (15B parameters) requires approximately 13.8 GB of VRAM with Q5_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 16 GB of VRAM.
Get started
— copy & paste to run locallyCopy-paste commands to run StarCoder2 15B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "bigcode/starcoder2-15b" \
--hf-file "starcoder2-15b-Q5_K_M.gguf" \
-c 4096 -ngl 99Quick specs
About this model
Related models
Quick picks
Best hardware
Run this model
Quantization options
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.9 GB | Low | — |
Q3_K_S | 3 | 7.4 GB | Low | — |
NVFP4 | 4 | 8.4 GB | Medium | — |
Q4_K_M | 4 | 9.2 GB | Medium | — |
Q5_K_M | 5 | 10.8 GB | High | — |
Q6_K | 6 | 12.3 GB | High | — |
Q8_0 | 8 | 16.1 GB | Very High | — |
F16 | 16 | 30.7 GB | Maximum | — |
Quality benchmarks
Coding
Reasoning
General
Source: official · 2024-02-28
Hardware compatibility
Computing compatibility...
Memory breakdown
Frequently asked questions
StarCoder2 15B (15B parameters) requires approximately 13.8 GB of VRAM with Q5_K_M quantization. Lower quantizations like Q4_K_M use less memory but may reduce quality.
Yes, RX 7600 XT 16GB can run StarCoder2 15B with a compatibility score of 50/100. It provides 16 GB of memory and achieves approximately 17.2 tokens per second.
The recommended quantization for StarCoder2 15B is Q5_K_M, which offers the best balance between model quality and memory efficiency. Higher quantizations preserve more quality but require more VRAM.
The top recommended hardware for StarCoder2 15B: RTX 4090 24GB (score: 57/100), RTX 5090 Laptop 24GB (score: 57/100), NVIDIA A30 24GB (score: 57/100). These provide the best combination of memory, bandwidth, and compute for running this model locally.
Yes, StarCoder2 15B is well-suited for coding. It was designed with these use cases in mind.
See also
